Fin vs Decagon

Actions

1. developerThe agent takes real actions through my APIs — refunds, order changes, subscription updates — with scoped auth per action

   weight 3 · round drawn
   Fin~partialclaimed6/10

   Fin's Procedures feature explicitly supports multi-step business logic with third-party systems (refunds, order changes, etc. are common examples in docs like 'Takes action to update external systems' and Procedures training), and data connectors (Stripe, Shopify) plus MCP/API access enable real API calls. However, evidence does not detail per-action scoped authentication/authorization — no documentation on granular auth scoping per action, permission boundaries, or audit trails for individual API calls. missing for 10: explicit scoped-auth-per-action documentation, permission/authorization model for individual actions, and independent verification that real refunds/subscription updates are executed successfully in production.

   • [claimed-docs] “Takes action to update external systems”
   • [claimed-docs] “Building Fin Procedures”
   • [claimed-docs] “Procedures let you train Fin to handle queries with multiple steps, business logic and third-party systems from start to finish.”
   • [claimed-docs] “Data connector templates are now generally available, with ready-to-install templates across Stripe, Shopify and Statuspage.”
   • [claimed-docs] “The Fin API platform includes APIs for answer generation and knowledge retrieval, available as a set or independently.”
   • [claimed-docs] “give you the tools to safely deploy AI within your own boundaries.”

   proof ↗⚑ flag

   Decagon~partialclaimed6/10

   Decagon documents scoped, short-lived JWT access for agent actions across customer systems and integrations 'with no custom code required,' plus explicit mention of agents adjusting 'refund logic,' supporting real action-taking with scoped auth. However there's no detailed documentation of per-action granularity (e.g., separate refund vs subscription-update scopes), no audit trail examples, and no independent/hands-on verification of actual API action execution — missing for 10: per-action scope definitions, action audit logging, and third-party verification of real transactional actions.

   • [claimed-docs] “Short-lived JWT tokens give AI agents real-time access to customer systems, scoped for minimal privilege and discarded after each session.”
   • [claimed-docs] “no custom code required”
   • [claimed-docs] “with no custom code required”
   • [claimed-docs] “Whether you’re refining tone, adjusting refund logic, or trialing a new onboarding experience, experimentation makes it safe to test, learn,…”
   • [claimed-docs] “lets your Decagon agent access and complete tasks inside systems where there isn’t a traditional integration available”

   proof ↗⚑ flag

Procedures

1. support ops leadI encode standard operating procedures the agent follows step-by-step for known issue types, with deterministic branching

   weight 2 · round to Fin
   Fin✓fullclaimed8/10

   Fin has a dedicated "Building Fin Procedures" feature explicitly for encoding multi-step SOPs with business logic and branching, plus "Giving Fin Guidance" and escalation-rule configuration to control deterministic branching to humans or other steps, and a testing suite to validate procedures before launch. This directly matches the story of encoding SOPs with deterministic branching for known issue types. Missing for 10: independent hands-on verification of branching logic depth/determinism and more detail on how conditional branches are authored (docs are descriptive marketing/help-article summaries rather than technical specs).

   • [claimed-docs] “Building Fin Procedures”
   • [claimed-docs] “Procedures let you train Fin to handle queries with multiple steps, business logic and third-party systems from start to finish.”
   • [claimed-docs] “Giving Fin Guidance”
   • [claimed-docs] “Manage Fin AI Agent's escalation guidance and rules”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation”
   • [claimed-docs] “Train Fin on your Procedures, knowledge, and policies, test performance before launch, deploy across every channel, then analyze and improve…”

   proof ↗⚑ flag

   Decagon✓fullclaimed7/10

   Decagon explicitly ships 'Agent Operating Procedures (AOPs)' described as letting teams 'define agent behavior in natural language, the same way you train human agents with SOPs,' with Duet auto-generating and iterating these procedures from real interactions. This directly matches encoding step-by-step SOPs for known issue types. Missing for 10: concrete technical detail on how branching logic/decision trees are structured and enforced deterministically, and independent/hands-on corroboration beyond vendor docs.

   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”
   • [claimed-docs] “Analyze past customer interactions to automatically generate Agent Operating Procedures (AOPs) rooted in actual customer needs.”
   • [claimed-docs] “Non-technical teams can architect and iterate on agent logic with Agent Operating Procedures (AOPs) and Duet”
   • [claimed-docs] “Non-technical teams can architect and iterate on agent logic with Agent Operating Procedures (AOPs) and Duet, while technical teams retain f…”
   • [claimed-docs] “As conversations reveal gaps and new opportunities, Duet automatically iterates on your agent, from the first launch through every improveme…”

   proof ↗⚑ flag

Agent access

1. ai-native userPoint an agent at llms.txt or agent-oriented docs

   weight 2 · round to Fin
   Fin✓fullprobed8/10

   A live probe confirms Fin publishes an llms.txt file at https://fin.ai/llms.txt (HTTP 200) explicitly framed to help LLMs understand its content structure, and Fin also offers an MCP server and a CLI agents can be pointed at for setup. Missing for 10: markdown-doc mirrors (docs-md probe 404) and a discoverable OpenAPI spec (all candidates 404), which would round out agent-friendly documentation.

   • [probe] “PROBE llms.txt: HTTP 200 at https://fin.ai/llms.txt # Fin — llms.txt > This file helps large language models (LLMs) understand the content …”
   • [claimed-docs] “Point your AI coding Agent at the Fin CLI to set up Fin on your website and configure your Intercom helpdesk — all without leaving your codi…”
   • [probe] “official MCP server documented at https://developers.intercom.com/docs/guides/mcp”
   • [probe] “PROBE docs-md: HTTP 404 at https://fin.ai/help/en/.md”
   • [probe] “PROBE openapi: all candidate paths 404 (https://fin.ai/openapi.json, https://fin.ai/swagger.json, https://fin.ai/api/openapi.json, https://f…”

   proof ↗⚑ flag

   Decagon~partialprobed6/10

   The probe confirms Decagon serves a working llms.txt file at decagon.ai/llms.txt (HTTP 200) with structured product description, showing the product is discoverable by AI agents pointed at agent-oriented docs. However, there's no evidence of broader agent-oriented documentation (e.g., structured API docs, OpenAPI spec which 404'd) or first-party guidance encouraging users to point agents at this file. missing for 10: OpenAPI/API-level machine-readable docs (404s confirmed), first-party documentation explicitly promoting llms.txt usage for AI agents, and independent corroboration of an agent successfully consuming the file.

   • [probe] “PROBE llms.txt: HTTP 200 at https://decagon.ai/llms.txt # Decagon Decagon is an enterprise-grade AI platform revolutionizing customer suppo…”
   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

2. ai-native userRun the product headlessly / in CI for automation

   weight 2 · round to Fin
   Fin~partialclaimed6/10

   Fin exposes a programmatic Agent API and Node/TypeScript SDK (intercom-fin-docs-3, intercom-fin-gh-3) that allows calling Fin from external code/services, which supports headless/CI-style automation, and webhooks (intercom-fin-docs-6) enable event-driven automation without a UI. However, there is no documented CI-specific tooling (e.g., a CLI test runner, GitHub Action, or exit-code-based automation harness) and the orchestration endpoints needed for autonomous agent-style headless runs are only in a Preview API (intercom-fin-docs-4). missing for 10: dedicated CI/CD integration or GitHub Action, documented headless test/run mode with exit codes, independent confirmation of automation reliability outside preview APIs.

   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [claimed-docs] “The new orchestration endpoints — Discover capabilities, Ask Fin, Run a procedure, and Escalate to a human — are currently available in the …”
   • [claimed-docs] “Webhooks allow you to subscribe to real-time notifications of events happening in Intercom; a Contact created, an incoming Conversation rece…”
   • [github] “The Intercom TypeScript library provides convenient access to the Intercom APIs from TypeScript.”

   proof ↗⚑ flag

   Decagon—none0/10

   Decagon is a SaaS conversational-AI platform for customer support with no-code configuration (AOPs, Duet, integrations 'no custom code required'); there is no documented CLI, headless mode, or CI-automation interface, and the openapi probe returned 404s, indicating no public API spec for automated/headless invocation. missing for 10: any CLI/SDK for headless execution, CI-pipeline integration docs, or public API reference enabling automation.

   • [claimed-docs] “no custom code required”
   • [claimed-docs] “with no custom code required”
   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

3. ai-native userPlug MCP servers into this product so it can use their tools

   weight 3 · round to Fin
   Fin✓fullclaimed7/10

   Fin's own docs state it 'works through MCP or API Data Connectors for your business tools' (docs-40/54) and explicitly describe connecting the 'Zapier MCP connector to Fin' (intdir-3), confirming Fin can consume external MCP servers as tools. Missing for 10: independent/hands-on verification of MCP tool usage in production and a broader list of supported MCP servers beyond the Zapier example.

   • [claimed-docs] “It works through MCP or API Data Connectors for your business tools, integrations with Intercom or your existing helpdesk”
   • [claimed-docs] “It works through MCP or API Data Connectors for your business tools, integrations with Intercom or your existing helpdesk, and native suppor…”
   • [claimed-docs] “Connect the Zapier MCP connector to Fin Manage workspace owners”
   • [claimed-docs] “Learn how to use the Model Context Protocol (MCP) to enable AI agents to securely access and interact with your Intercom data whenever helpf…”
   • [claimed-docs] “MCP is a protocol that enables AI tools and applications to securely access and interact with your Intercom data whenever helpful.”

   proof ↗⚑ flag

   Decagon~partialclaimed3/10

   Decagon has a blog post explicitly about MCP ("getting the most out of MCP") describing an infrastructure layer to curate, scope, and evaluate tool usage, indicating some MCP integration capability, but there is no concrete documentation of how to actually plug an MCP server into an agent, no config steps, and no independent corroboration of it working. missing for 10: technical setup docs for adding an MCP server, list of supported MCP servers/tools, hands-on or independent verification that agents actually invoke MCP tools.

   • [claimed-docs] “Bridging that gap requires an infrastructure layer purpose-built to curate, scope, and evaluate how tools are actually used.”

   proof ↗⚑ flag

4. ai-native userConnect an agent via an official MCP server

   weight 3 · round to Fin
   Fin✓fullprobed8/10

   Intercom publishes official docs for an MCP server that lets AI agents securely access and interact with Intercom data, and this is corroborated by a dedicated probe confirming the documented endpoint plus Fin's own integrations page referencing MCP connectivity. missing for 10: independent/hands-on confirmation of the MCP server working in practice beyond vendor docs, and clearer detail on scope/auth setup.

   • [claimed-docs] “Learn how to use the Model Context Protocol (MCP) to enable AI agents to securely access and interact with your Intercom data whenever helpf…”
   • [claimed-docs] “MCP is a protocol that enables AI tools and applications to securely access and interact with your Intercom data whenever helpful.”
   • [claimed-docs] “It works through MCP or API Data Connectors for your business tools, integrations with Intercom or your existing helpdesk”
   • [claimed-docs] “It works through MCP or API Data Connectors for your business tools, integrations with Intercom or your existing helpdesk, and native suppor…”
   • [probe] “official MCP server documented at https://developers.intercom.com/docs/guides/mcp”

   proof ↗⚑ flag

   Decagon—none0/10

   Decagon's MCP blog post (decagon-docs-3) discusses using MCP to curate/scope tool access for its own agents (i.e., Decagon as an MCP client consuming external tools), not exposing an official MCP server that lets an external AI agent connect into Decagon. No documentation, endpoint, or announcement of a first-party Decagon MCP server was found, and the OpenAPI/spec probes returned 404s with no MCP-specific server mentioned.

   • [claimed-docs] “Bridging that gap requires an infrastructure layer purpose-built to curate, scope, and evaluate how tools are actually used.”
   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

5. ai-native userDrive the product through a documented public API

   weight 3 · round to Fin
   Fin~partialprobed7/10

   Fin ships a documented Fin Agent API and an official TypeScript SDK (intercom-node) with concrete usage examples, plus a separate API platform for answer generation/knowledge retrieval, giving AI-native users a real programmatic path to drive Fin. However, key orchestration endpoints (Discover capabilities, Ask Fin, Run a procedure, Escalate) are only in Preview API version, and automated probes found no discoverable OpenAPI/swagger spec, suggesting the public API surface is not fully standardized/discoverable yet. Missing for 10: stable (non-preview) orchestration endpoints, a machine-readable OpenAPI spec, and independent hands-on developer corroboration beyond vendor docs/SDK repo.

   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [claimed-docs] “The new orchestration endpoints — Discover capabilities, Ask Fin, Run a procedure, and Escalate to a human — are currently available in the …”
   • [claimed-docs] “The Fin API platform includes APIs for answer generation and knowledge retrieval, available as a set or independently.”
   • [claimed-docs] “Build world-class customer experiences, agents, and products with the best-performing models for customer service.”
   • [github] “await client.aiContent.createContentImportSource({ url: "https://www.example.com" });”
   • [github] “List endpoints are paginated. The SDK provides an iterator so that you can simply loop over the items”
   • [github] “The Intercom TypeScript library provides convenient access to the Intercom APIs from TypeScript.”
   • [probe] “PROBE openapi: all candidate paths 404 (https://fin.ai/openapi.json, https://fin.ai/swagger.json, https://fin.ai/api/openapi.json, https://f…”

   proof ↗⚑ flag

   Decagon—none0/10

   No evidence of a documented public API; the openapi probe returned 404s across all candidate paths and no docs reference an API reference, SDK, or programmatic endpoint. Integrations are described as 'no custom code required' (docs-19, docs-26), suggesting no-code/UI-driven configuration rather than a documented API for AI-native control.

   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”
   • [claimed-docs] “no custom code required”
   • [claimed-docs] “with no custom code required”

   proof ↗⚑ flag

6. ai-native userIssue scoped/least-privilege API credentials for an agent

   weight 2 · round to Decagon
   Fin—none0/10

   Evidence covers API access, MCP, webhooks, and integrations, but there is no mention of scoped or least-privilege API credential/token issuance for agents (e.g., granular permission scopes, per-agent API keys). Absence of evidence for an applicable capability yields 'none'. missing for 10: scoped/least-privilege credential issuance, API key/token permission granularity, agent-specific credential management docs.

   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [claimed-docs] “Learn how to use the Model Context Protocol (MCP) to enable AI agents to securely access and interact with your Intercom data whenever helpf…”
   • [claimed-docs] “MCP is a protocol that enables AI tools and applications to securely access and interact with your Intercom data whenever helpful.”

   proof ↗⚑ flag

   Decagon~partialclaimed4/10

   Decagon documents short-lived, scoped JWT tokens for agent access to customer systems and identity-provider (Okta/Entra) integration, which shows some least-privilege credential mechanics, but this is about the agent's own runtime access rather than a user-facing capability to explicitly issue/manage scoped API credentials for an agent. There's no evidence of an API/console feature letting an AI-native user provision, scope, or revoke discrete credentials themselves. missing for 10: user-facing credential issuance/management UI or API, granular scoping controls exposed to users, independent verification of the JWT scoping claims.

   • [claimed-docs] “Short-lived JWT tokens give AI agents real-time access to customer systems, scoped for minimal privilege and discarded after each session.”
   • [claimed-docs] “Decagon integrates with identity providers like Okta and Microsoft Entra, enabling secure access across systems without additional passwords…”

   proof ↗⚑ flag

7. ai-native userBuild against official SDKs

   weight 2 · round to Fin
   Fin~partialprobed6/10

   Intercom publishes an official TypeScript/Node SDK (intercom-node) with convenient API access and iterator support, plus a documented Fin Agent API and API platform for programmatic use. However, only one official SDK language is evidenced, there's no public OpenAPI spec (probe returned 404s across all candidate paths), and no independent corroboration of SDK quality/adoption exists. Missing for 10: multi-language SDK coverage, publicly discoverable OpenAPI/schema, and third-party validation of SDK reliability.

   • [github] “await client.aiContent.createContentImportSource({ url: "https://www.example.com" });”
   • [github] “List endpoints are paginated. The SDK provides an iterator so that you can simply loop over the items”
   • [github] “The Intercom TypeScript library provides convenient access to the Intercom APIs from TypeScript.”
   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [claimed-docs] “The Fin API platform includes APIs for answer generation and knowledge retrieval, available as a set or independently.”
   • [probe] “PROBE openapi: all candidate paths 404 (https://fin.ai/openapi.json, https://fin.ai/swagger.json, https://fin.ai/api/openapi.json, https://f…”

   proof ↗⚑ flag

   Decagon—none0/10

   No evidence of official SDKs for developers; the pack shows only no-code integrations, MCP blog commentary, and a failed openapi probe (404s), with no SDK documentation or API libraries surfaced.

   • [claimed-docs] “Bridging that gap requires an infrastructure layer purpose-built to curate, scope, and evaluate how tools are actually used.”
   • [claimed-docs] “no custom code required”
   • [claimed-docs] “with no custom code required”
   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

8. ai-native userSubscribe to events via webhooks

   weight 2 · round to Fin
   Fin✓fullclaimed8/10

   Intercom's developer docs explicitly document webhooks for subscribing to real-time event notifications (Contact created, Conversation received, Outbound Message receipt), directly matching the story. Missing for 10: independent/hands-on corroboration of webhook reliability and broader event coverage beyond the three examples cited.

   • [claimed-docs] “Webhooks allow you to subscribe to real-time notifications of events happening in Intercom; a Contact created, an incoming Conversation rece…”
   • [claimed-docs] “Webhooks allow you to subscribe to real-time notifications of events happening in Intercom”

   proof ↗⚑ flag

   Decagon—none0/10

   No evidence in the pack mentions webhooks, event subscriptions, or any push-notification mechanism; the OpenAPI probe returned 404s and no API reference documenting webhook endpoints was found. This axis applies (Decagon integrates with external systems and could plausibly offer webhooks) but there is no supporting evidence.

   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

Agentic features

1. ai-native userGet AI-generated insights and suggestions from my data inside the product

   weight 2 · round to Decagon
   Fin~partialclaimed6/10

   Fin's 'Insights', 'AI Recommendations', 'AI Topics', 'Trends', and CX Score features (docs-16, docs-47, docs-61) show it surfaces AI-generated analysis of conversation data, and 'Monitors'/'Custom AI Scorecards' extend this to proactive alerting. However this is framed around support-conversation analytics rather than a general 'insights from your data' experience, and there's no independent/hands-on validation of accuracy or usefulness of these AI-generated insights. missing for 10: independent corroboration of insight quality, broader data-source coverage beyond support conversations, concrete UI examples of AI-generated suggestions.

   • [claimed-docs] “Operator, CX Score, AI Topics, Trends, AI Recommendations, Monitors, Custom AI Scorecards, Incident Detection”
   • [claimed-docs] “Insights continuously analyzes every Fin and human conv”
   • [claimed-docs] “With Insights, Monitors New, and Recommendations, you have everything you need to understand, measure, and improve your customer experience”
   • [claimed-docs] “Train Fin on your Procedures, knowledge, and policies, test performance before launch, deploy across every channel, then analyze and improve…”

   proof ↗⚑ flag

   Decagon✓fullclaimed8/10

   Decagon's Insights & Reporting and Suggestions features let users ask open-ended natural-language questions (e.g. 'Why are customers requesting refunds?') and get AI-driven analysis, heatmaps, and auto-generated content drafts based on detected knowledge gaps, directly surfacing AI-generated insights from customer data. Duet further auto-generates Agent Operating Procedures from past interactions and iterates based on conversation patterns. Missing for 10: independent/hands-on validation of insight quality and no detail on underlying analytics accuracy or limitations.

   • [claimed-docs] “Ask open-ended questions like “Why are customers requesting refunds?” to instantly analyze customer conversations.”
   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues, so your…”
   • [claimed-docs] “Customizable visual heatmaps... Quickly identify spikes or dips in key metrics.”
   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues”
   • [claimed-docs] “Ask AI powers deeper analysis through flexible, conversational queries.”
   • [claimed-docs] “Analyze past customer interactions to automatically generate Agent Operating Procedures (AOPs) rooted in actual customer needs.”
   • [claimed-docs] “As conversations reveal gaps and new opportunities, Duet automatically iterates on your agent, from the first launch through every improveme…”

   proof ↗⚑ flag

2. ai-native userSet up automations that run autonomously in the background

   weight 2 · round drawn
   Fin~partialcommunity6/10

   Fin's Procedures, Tasks, Workflows, and Proactive Outbound Suite are documented as letting Fin autonomously handle multi-step customer queries, business logic, and third-party system actions end-to-end without human intervention, and Fin resolves a majority of queries (76%) without agent involvement. However, evidence is entirely vendor marketing/help docs with no independent or hands-on confirmation of scheduling/background execution mechanics, and one community comment questions whether open-ended autonomous agent behavior is even desirable versus deterministic workflows. Missing for 10: independent verification of autonomous background execution, technical detail on triggers/scheduling for Procedures/Tasks, and hands-on confirmation that these run without ongoing human oversight.

   • [claimed-docs] “Building Fin Procedures”
   • [claimed-docs] “Manage your Fin workflows”
   • [claimed-docs] “Introduction to Fin Tasks”
   • [claimed-docs] “Procedures let you train Fin to handle queries with multiple steps, business logic and third-party systems from start to finish.”
   • [claimed-docs] “Proactive Outbound Suite”
   • [claimed-docs] “Fin resolves 76% of customer queries on average, and handles multi-step queries end to end. Your customers get fast, accurate answers. Your …”
   • [community] “A commenter with a decade of customer-service AI/NLP experience argued 99% of support tasks (order status, returns, troubleshooting) are ful…”

   proof ↗⚑ flag

   Decagon~partialclaimed6/10

   Decagon's Proactive Agents can carry context across sessions and autonomously initiate calls/recommendations based on customer signals, and Watchtower continuously monitors every interaction against custom criteria without manual triggering — both indicate background/autonomous operation. However, there's no explicit documentation of a general-purpose automation/scheduling framework (e.g., triggers, cron-like workflows, or arbitrary background tasks) beyond these two specific features. Missing for 10: a dedicated automation/scheduler product surface, independent/hands-on verification of autonomous behavior, and broader configurability beyond proactive outreach and monitoring.

   • [claimed-docs] “Carry conversation history across sessions to pick up where you left off and make proactive recommendations based on customer signals.”
   • [claimed-docs] “Initiate intelligent, on-brand calls that reliably reach customers at the right moments throughout their journey.”
   • [claimed-docs] “Maintain customer context across sessions, complementing your systems of record with data portability and control.”
   • [claimed-docs] “Whether you're monitoring for compliance risks, negative sentiment, or upsell signals, Watchtower reviews every interaction against your cus…”

   proof ↗⚑ flag

3. ai-native userDelegate tasks to a built-in AI assistant inside the product

   weight 3 · round to Fin
   Fin✓fullcommunity8/10

   Fin is itself a built-in AI agent that customers and support teams delegate tasks to — it resolves ~76% of queries end-to-end, handles multi-step 'Procedures' with business logic, takes actions on external systems, and works across channels (chat, email, voice, Slack, WhatsApp). This is documented extensively in first-party docs/help articles covering delegation-style use (Fin Tasks, Fin Procedures, escalation rules, API/Agent API for programmatic delegation). missing for 10: independent hands-on validation of delegation quality (one HN commenter says they've never seen Fin in the wild, and another questions whether autonomous agent reasoning is even the right approach vs deterministic workflows), and no live product screenshots/demo confirming smooth end-user delegation experience.

   • [claimed-docs] “Handles tickets, cases, emails, live chat, WhatsApp, SMS, and more”
   • [claimed-docs] “Takes action to update external systems”
   • [claimed-docs] “Building Fin Procedures”
   • [claimed-docs] “Introduction to Fin Tasks”
   • [claimed-docs] “Procedures let you train Fin to handle queries with multiple steps, business logic and third-party systems from start to finish.”
   • [claimed-docs] “Fin resolves 76% of customer queries on average, and handles multi-step queries end to end. Your customers get fast, accurate answers. Your …”
   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [community] “"I've never seen or interacted with Fin that I've noticed... I suspect it's very visible for Intercom, but not necessarily so for everyone e…”
   • [community] “A commenter with a decade of customer-service AI/NLP experience argued 99% of support tasks (order status, returns, troubleshooting) are ful…”

   proof ↗⚑ flag

   Decagon~partialclaimed5/10

   Decagon's Duet acts as an in-product AI helper that lets (non-technical) users delegate specific tasks — auto-generating Agent Operating Procedures from past conversations, drafting tests, and producing content suggestions — which is a form of delegating work to a built-in assistant, but it's narrowly scoped to agent-configuration/QA tasks rather than a general-purpose assistant for arbitrary user tasks. missing for 10: evidence of a general-purpose in-product assistant handling open-ended user requests beyond AOP/test/content generation, independent/hands-on validation of Duet's delegation quality, and detail on how broadly tasks can be delegated versus templated workflows.

   • [claimed-docs] “Non-technical teams can architect and iterate on agent logic with Agent Operating Procedures (AOPs) and Duet”
   • [claimed-docs] “Analyze past customer interactions to automatically generate Agent Operating Procedures (AOPs) rooted in actual customer needs.”
   • [claimed-docs] “Use Duet to easily generate tests covering diverse pathways that verify whether agents respond accurately, follow policies, and reflect your…”
   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues, so your…”

   proof ↗⚑ flag

4. ai-native userOperate the product with natural-language commands

   weight 2 · round drawn
   Fin~partialclaimed6/10

   Fin's core function is responding to customer queries in natural language across channels, and admins can 'give Fin guidance' and train Procedures using natural-language instructions (docs-22, docs-44, docs-60); the Fin CLI also lets an AI coding agent configure Fin setup based on natural-language prompts (docs-38). However there's no clear evidence of a full natural-language command interface for operating/administering the product itself (e.g., configuring settings, running reports) beyond training/guidance content. Missing for 10: documented NL-driven admin console commands, evidence of broad NL task execution beyond content/guidance training, and independent verification of this capability in practice.

   • [claimed-docs] “Giving Fin Guidance”
   • [claimed-docs] “Procedures let you train Fin to handle queries with multiple steps, business logic and third-party systems from start to finish.”
   • [claimed-docs] “Point your AI coding Agent at the Fin CLI to set up Fin on your website and configure your Intercom helpdesk — all without leaving your codi…”
   • [claimed-docs] “Train Fin on your knowledge, data, policies, and tone of voice. Fin will use everything it knows to deliver consistent, accurate answers”

   proof ↗⚑ flag

   Decagon~partialclaimed6/10

   Decagon lets operators define agent behavior, flagging criteria, and analytics queries via natural language (AOPs, Watchtower, Ask AI), which supports natural-language operation of the product for configuration/analysis purposes. However, this is primarily aimed at business/support-ops users configuring an agent, not a general 'AI-native user' issuing arbitrary natural-language commands to control the whole product end-to-end. missing for 10: evidence of natural-language command interface for broader product operations (e.g., admin tasks, integrations setup, deployment) beyond AOPs/Watchtower/Insights, and independent/hands-on verification of this capability.

   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Ask AI powers deeper analysis through flexible, conversational queries.”
   • [claimed-docs] “Ask open-ended questions like “Why are customers requesting refunds?” to instantly analyze customer conversations.”

   proof ↗⚑ flag

Api quality

1. ai-native userExplore an interactive API reference with runnable examples

   weight 2 · round drawn
   Fin—none0/10

   While Fin has API docs (developers.intercom.com) and an SDK on GitHub, there is no evidence of an interactive API reference with runnable/try-it examples; a probe explicitly found no OpenAPI/Swagger spec published at any candidate path, and no docs mention a live API console.

   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [probe] “PROBE openapi: all candidate paths 404 (https://fin.ai/openapi.json, https://fin.ai/swagger.json, https://fin.ai/api/openapi.json, https://f…”
   • [github] “The Intercom TypeScript library provides convenient access to the Intercom APIs from TypeScript.”

   proof ↗⚑ flag

   Decagon—none0/10

   No evidence of an interactive API reference or runnable examples; the openapi probe returned 404 across all candidate paths and no docs mention API documentation with runnable examples. Missing for 10: any API reference page, SDK/runnable code samples, interactive docs like Swagger/Redoc, or developer sandbox.

   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

2. ai-native userDownload a machine-readable API spec (OpenAPI or equivalent)

   weight 2 · round drawn
   Fin—none0/10

   Fin has documented REST APIs (Fin Agent API, Node/TS SDK) but no evidence of a downloadable machine-readable spec; a direct probe for OpenAPI/swagger files at common paths returned 404 for all candidates, and no docs page links such a spec.

   • [probe] “PROBE openapi: all candidate paths 404 (https://fin.ai/openapi.json, https://fin.ai/swagger.json, https://fin.ai/api/openapi.json, https://f…”
   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [claimed-docs] “The new orchestration endpoints — Discover capabilities, Ask Fin, Run a procedure, and Escalate to a human — are currently available in the …”

   proof ↗⚑ flag

   Decagon—none0/10

   Probe explicitly checked common OpenAPI/swagger endpoints and all returned 404, and no documentation references a downloadable machine-readable API spec; only an llms.txt file was found, which is not an API spec.

   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”
   • [probe] “PROBE llms.txt: HTTP 200 at https://decagon.ai/llms.txt # Decagon Decagon is an enterprise-grade AI platform revolutionizing customer suppo…”

   proof ↗⚑ flag

3. ai-native userTest against a sandbox environment without touching production data

   weight 1 · round to Decagon
   Fin~partialclaimed4/10

   Fin documents a testing/preview workflow ('How to preview and test Fin', testing across thousands of scenarios before going live) that implies an isolated test/preview mode, but no evidence explicitly describes a sandbox environment separate from production data or confirms production data isolation during testing. Missing for 10: explicit sandbox/staging environment documentation, confirmation that test scenarios don't touch live customer data, and any independent/hands-on corroboration of this isolation.

   • [claimed-docs] “How to preview and test Fin”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation... …”

   proof ↗⚑ flag

   Decagon~partialclaimed5/10

   Decagon's Testing & QA product describes 'Simulations' that validate agent behavior 'before deploying to production and with every subsequent update' (decagon-docs-29) and Duet-generated test suites (decagon-docs-11), implying a pre-production testing environment. However, there is no explicit documentation of a dedicated sandbox with isolated/non-production data, and the Experiments feature explicitly runs live in production (decagon-docs-7), which cuts against a clear sandbox-vs-production separation. Missing for 10: explicit description of sandbox data isolation, confirmation that test/simulation environments don't touch live customer data, and independent verification of this claim.

   • [claimed-docs] “Simulations, Decagon's integrated testing suite, help teams validate agent behavior across channels before deploying to production and with …”
   • [claimed-docs] “Use Duet to easily generate tests covering diverse pathways that verify whether agents respond accurately, follow policies, and reflect your…”
   • [claimed-docs] “Define your variable and traffic split, then measure results against a stable control group directly in production.”

   proof ↗⚑ flag

4. ai-native userRely on versioned APIs with a documented deprecation policy

   weight 2 · round drawn
   Fin—none0/10

   The evidence shows Fin has a REST API with a 'Preview API version' for new endpoints (docs-4) and a general changes/changelog page (docs-1), but there is no documented versioning scheme or explicit deprecation policy (e.g., version sunset timelines, backward-compatibility guarantees) cited anywhere in the pack.

   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [claimed-docs] “The new orchestration endpoints — Discover capabilities, Ask Fin, Run a procedure, and Escalate to a human — are currently available in the …”
   • [claimed-docs] “See exactly when errors started, which status codes came back, and confirm a fix actually worked.”

   proof ↗⚑ flag

   Decagon—none0/10

   No evidence of a public/versioned API, API changelog, or deprecation policy; the openapi probe returned 404s across all candidate paths and no docs mention API versioning or deprecation practices.

   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

1. ai-native userPerform bulk operations across many items at once

   weight 2 · round to Fin
   Fin~partialclaimed3/10

   Fin exposes a programmatic API and TypeScript SDK with a pagination iterator for looping over items, plus a content-import endpoint, implying some capacity for scripted bulk actions, but there is no documented bulk-edit, bulk-resolve, or batch-action feature for AI-native users. missing for 10: explicit bulk operation/batch endpoint documentation, hands-on evidence of processing many items in one call, and any UI/CLI bulk-action support.

   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [github] “await client.aiContent.createContentImportSource({ url: "https://www.example.com" });”
   • [github] “List endpoints are paginated. The SDK provides an iterator so that you can simply loop over the items”

   proof ↗⚑ flag

   Decagon—none0/10

   No evidence describes a bulk-operations feature (e.g., batch editing, bulk tagging, bulk export/import of tickets or conversations) for AI-native users; the closest items describe aggregate analysis (Watchtower reviewing every interaction, Insights analyzing many conversations) rather than user-driven bulk actions across items.

   ⚑ flag

2. ai-native userDefine rules that trigger actions automatically on events

   weight 3 · round to Fin
   Fin~partialclaimed6/10

   Fin exposes several automation primitives that fit an event→action rule model: webhooks that fire on real-time Intercom events, configurable escalation guidance/rules, Fin Procedures/workflows, and Operator/Monitors for incident-triggered behavior. This is more than none, but the pack lacks a first-class 'if event X then action Y' rule-builder walkthrough or independent confirmation of how flexible/robust these triggers are. Missing for 10: a documented dedicated rules/automation builder UI, concrete example of a user-defined trigger-condition-action rule, and independent (non-vendor) evidence that automation rules work reliably in practice.

   • [claimed-docs] “Webhooks allow you to subscribe to real-time notifications of events happening in Intercom; a Contact created, an incoming Conversation rece…”
   • [claimed-docs] “Manage Fin AI Agent's escalation guidance and rules”
   • [claimed-docs] “Building Fin Procedures”
   • [claimed-docs] “Manage your Fin workflows”
   • [claimed-docs] “Operator, CX Score, AI Topics, Trends, AI Recommendations, Monitors, Custom AI Scorecards, Incident Detection”

   proof ↗⚑ flag

   Decagon~partialclaimed5/10

   Decagon offers Watchtower, which monitors conversations against natural-language criteria and can flag/surface events (compliance risks, sentiment, etc.), and Proactive Agents that act on customer signals (e.g., initiating calls, recommendations) — both function as event-triggered automation. However, there is no explicit documentation of a general-purpose rule-builder (if-event-then-action) framework that an AI-native user could configure directly; the evidence describes narrower, product-specific triggering mechanisms rather than a flexible automation-rules engine. Missing for 10: explicit rule/condition builder UI or API, documentation of arbitrary event types and action bindings, and independent verification of how these triggers are configured.

   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Whether you're monitoring for compliance risks, negative sentiment, or upsell signals, Watchtower reviews every interaction against your cus…”
   • [claimed-docs] “Carry conversation history across sessions to pick up where you left off and make proactive recommendations based on customer signals.”
   • [claimed-docs] “Initiate intelligent, on-brand calls that reliably reach customers at the right moments throughout their journey.”
   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”

   proof ↗⚑ flag

3. ai-native userSchedule recurring jobs or workflows

   weight 2 · round drawn
   Fin—none0/10

   Fin's evidence covers webhooks (event-driven, not scheduled), procedures, and a Proactive Outbound Suite, but nothing describes cron-like scheduling of recurring jobs or workflows on a time-based schedule. missing for 10: any documentation of scheduled/recurring workflow triggers, cron-style job scheduling, or recurring automation configuration.

   • [claimed-docs] “Webhooks allow you to subscribe to real-time notifications of events happening in Intercom; a Contact created, an incoming Conversation rece…”
   • [claimed-docs] “Proactive Outbound Suite”
   • [claimed-docs] “Procedures let you train Fin to handle queries with multiple steps, business logic and third-party systems from start to finish.”

   proof ↗⚑ flag

   Decagon—none0/10

   Decagon's evidence covers customer-support agent features (chat, voice, email, analytics, testing, proactive outreach) but nothing addresses scheduling recurring jobs or workflows in the AI-native/automation-depth sense; proactive agents mention initiating calls at 'right moments' but no scheduling/cron-like mechanism is documented. missing for 10: any scheduler, cron/recurring trigger config, or workflow automation timing controls.

   ⚑ flag

4. ai-native userVersion, review, and roll back my automations

   weight 1 · round to Decagon
   Fin—none0/10

   The evidence pack shows Fin Procedures/workflow management and testing before go-live ('test it across thousands of scenarios before anything goes live, roll out changes with control'), but there is no concrete documentation of version history, diffing/review of automation changes, or an explicit rollback mechanism for Procedures/Workflows/Guidance. Nothing describes a changelog, version comparison, or a 'revert to previous version' feature for automations.

   • [claimed-docs] “Building Fin Procedures”
   • [claimed-docs] “Manage your Fin workflows”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation”

   proof ↗⚑ flag

   Decagon~partialclaimed4/10

   Decagon offers some adjacent controls—technical teams retain 'visibility and control over guardrails, integrations, and versioning' and Simulations/testing validate agent behavior 'before deploying to production and with every subsequent update'—suggesting some change-management workflow exists, but there's no explicit documentation of version history browsing, diffing, review/approval workflows, or a rollback mechanism to a prior automation state. missing for 10: explicit versioning UI/history, review/approval workflow for changes, documented rollback mechanism, independent confirmation of these capabilities in use.

   • [claimed-docs] “technical teams retain full visibility and control over guardrails, integrations, and versioning”
   • [claimed-docs] “Non-technical teams can architect and iterate on agent logic with Agent Operating Procedures (AOPs) and Duet, while technical teams retain f…”
   • [claimed-docs] “Simulations, Decagon's integrated testing suite, help teams validate agent behavior across channels before deploying to production and with …”

   proof ↗⚑ flag

Channels

1. support leaderOne agent covers chat, email, and in-app, plus the channels my customers actually use — Slack, WhatsApp, social

   weight 2 · round to Fin
   Fin✓fullclaimed7/10

   Fin explicitly documents multi-channel support spanning live chat/messenger (in-app), email, Slack, WhatsApp, SMS, and voice, with dedicated channel pages for each (docs-9, docs-50, docs-51, docs-65, docs-48). Missing for 10: explicit named coverage of broader 'social' channels like Instagram/Facebook/X/Twitter — only 'and more' is implied, not documented by name, so the specific 'social' claim in the story is only partially evidenced.

   • [claimed-docs] “Handles tickets, cases, emails, live chat, WhatsApp, SMS, and more”
   • [claimed-docs] “Fin delivers unrivaled performance across every channel. Wherever customers reach out, Fin provides fast, accurate, personalized support”
   • [claimed-docs] “Fin over Email resolves everything from simple FAQs to complex threads with attachments and multiple recipients”
   • [claimed-docs] “Fin for Slack supports your customers directly in Slack communities and dedicated Connect channels, resolving questions instantly with repli…”
   • [claimed-docs] “Fin for messenger brings the highest-performing AI Agent for customer service to live chat, delivering exceptional, on-brand support”

   proof ↗⚑ flag

   Decagon~partialclaimed5/10

   Decagon clearly offers a unified agent across chat (web, mobile, messaging platforms), email, and voice, suggesting broad channel reach [decagon-docs-13][decagon-docs-14][decagon-docs-15], with proactive/session-continuity features tying channels together [decagon-docs-22][decagon-docs-28]. However, the evidence never names specific channels like Slack, WhatsApp, or social media explicitly — only generic 'messaging platforms' is mentioned. Missing for 10: explicit documentation naming Slack, WhatsApp, and social media integrations as supported channels, plus any customer proof of omnichannel handoff across these specific channels.

   • [claimed-docs] “Flexible, on-brand chat built for conversations across web, mobile, and messaging platforms.”
   • [claimed-docs] “Voice AI agents built for natural, multilingual conversations and customized to your brand.”
   • [claimed-docs] “Intelligent email that understands context, stays on brand, and handles complex customer threads.”
   • [claimed-docs] “Carry conversation history across sessions to pick up where you left off and make proactive recommendations based on customer signals.”
   • [claimed-docs] “Maintain customer context across sessions, complementing your systems of record with data portability and control.”

   proof ↗⚑ flag

Languages

1. support leaderThe agent supports customers in many languages, even where my knowledge base exists only in English

   weight 2 · round to Fin
   Fin✓fullclaimed7/10

   Intercom/Fin explicitly documents multi-language support ('Use Fin AI Agent in multiple languages') allowing Fin to answer in customers' languages even when the knowledge base content is authored in English, which directly matches the story. Missing for 10: independent/hands-on verification of translation quality or explicit detail on how English-only knowledge base content is translated/handled per language.

   • [claimed-docs] “Use Fin AI Agent in multiple languages”

   proof ↗⚑ flag

   Decagon~partialclaimed4/10

   Only the Voice product page explicitly claims multilingual capability ('built for natural, multilingual conversations'); there is no evidence that Chat or Email channels support multiple languages, nor any documentation of translating or drawing from an English-only knowledge base to serve other languages. Missing for 10: explicit multilingual support claims for chat/email channels, description of how KB content is translated/localized, and independent verification of multilingual quality.

   • [claimed-docs] “Voice AI agents built for natural, multilingual conversations and customized to your brand.”
   • [claimed-docs] “Flexible, on-brand chat built for conversations across web, mobile, and messaging platforms.”
   • [claimed-docs] “Intelligent email that understands context, stays on brand, and handles complex customer threads.”

   proof ↗⚑ flag

Voice

1. support leaderThe agent handles phone calls — speech in, speech out — with the same knowledge and actions as chat

   weight 2 · round to Fin
   Fin✓fullclaimed7/10

   Fin Voice is a documented dedicated product ('Deploy Fin Voice', 'Fin Voice 2 runs on Apex Flash... latency-sensitive tasks', 'grounded in your knowledge base, and trained to apply your specific policies on every call'), indicating speech-in/speech-out phone support using the same knowledge base as chat. Missing for 10: independent/hands-on validation of voice call quality or action-taking parity specifically for phone, and no detail on how actions (e.g., updating systems) work identically on voice vs chat.

   • [claimed-docs] “Deploy Fin Voice”
   • [claimed-docs] “grounded in your knowledge base, and trained to apply your specific policies on every call.”
   • [claimed-docs] “Fin Voice 2 runs on Apex Flash - our latest Fin model, built for latency-sensitive tasks.”
   • [claimed-docs] “Fin delivers unrivaled performance across every channel. Wherever customers reach out, Fin provides fast, accurate, personalized support”

   proof ↗⚑ flag

   Decagon~partialclaimed6/10

   Decagon explicitly offers a Voice product for 'natural, multilingual conversations' and can 'initiate intelligent, on-brand calls,' and its platform architecture (AOPs, integrations, guardrails) is shared across channels including chat, implying shared knowledge/actions. However, there's no explicit vendor statement confirming full parity of knowledge/actions between voice and chat, no technical detail on speech-to-speech quality, and no independent/hands-on corroboration of call handling. Missing for 10: explicit parity confirmation between voice and chat agent logic, technical/latency details of the voice pipeline, independent or customer testimonial evidence of voice call handling in production.

   • [claimed-docs] “Voice AI agents built for natural, multilingual conversations and customized to your brand.”
   • [claimed-docs] “Initiate intelligent, on-brand calls that reliably reach customers at the right moments throughout their journey.”
   • [claimed-docs] “Flexible, on-brand chat built for conversations across web, mobile, and messaging platforms.”
   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”
   • [claimed-docs] “Non-technical teams can architect and iterate on agent logic with Agent Operating Procedures (AOPs) and Duet, while technical teams retain f…”

   proof ↗⚑ flag

Handoff

1. support leaderWhen the agent escalates, the human gets the full conversation, a summary, and collected details — the customer never repeats themselves

   weight 3 · round to Fin
   Fin~partialclaimed6/10

   Docs confirm Fin can transfer/escalate to human agents within the preferred inbox, has configurable escalation guidance/rules, and supports hand-off via JavaScript and an orchestration 'Escalate to a human' API — showing escalation handoff is a real, built feature. However, none of the evidence explicitly confirms that a generated summary or a structured collection of gathered details is passed along with the full conversation to the human agent, so the 'never repeats themselves' guarantee is not directly documented. Missing for 10: explicit documentation of automatic conversation summary generation at escalation, evidence that collected customer details are packaged and handed off, and independent/hands-on confirmation the handoff actually eliminates repetition.

   • [claimed-docs] “The new orchestration endpoints — Discover capabilities, Ask Fin, Run a procedure, and Escalate to a human — are currently available in the …”
   • [claimed-docs] “Transfers to agents directly in preferred Inbox”
   • [claimed-docs] “Manage Fin AI Agent's escalation guidance and rules”
   • [claimed-docs] “Hand-off with JavaScript”

   proof ↗⚑ flag

   Decagon—none0/10

   The evidence pack contains no mention of escalation-to-human handoff, conversation summaries handed to agents, or collected-details transfer preventing repetition. Related items about cross-session context (decagon-docs-22, decagon-docs-28) address agent-to-customer continuity, not human-agent handoff, so the specific story is unevidenced despite being a plausible capability for a support AI platform.

   ⚑ flag

Rules

1. support ops leadI configure when the agent must hand off — by topic, sentiment, customer tier, or explicit request — and it reliably obeys

   weight 2 · round to Fin
   Fin~partialclaimed6/10

   Docs explicitly cover configuring escalation/handoff behavior — 'Manage Fin AI Agent's escalation guidance and rules', 'Giving Fin Guidance', 'Hand-off with JavaScript', an orchestration 'Escalate to a human' API endpoint, and 'Transfers to agents directly in preferred Inbox' — showing topic/guidance-based and explicit-request handoff configuration exists. However, none of the evidence specifically documents sentiment- or customer-tier-based handoff triggers, nor is there independent/hands-on verification that the agent reliably obeys these rules in practice. Missing for 10: explicit sentiment/tier-based escalation config docs, and third-party evidence confirming reliable adherence to configured handoff rules.

   • [claimed-docs] “Manage Fin AI Agent's escalation guidance and rules”
   • [claimed-docs] “Giving Fin Guidance”
   • [claimed-docs] “Hand-off with JavaScript”
   • [claimed-docs] “The new orchestration endpoints — Discover capabilities, Ask Fin, Run a procedure, and Escalate to a human — are currently available in the …”
   • [claimed-docs] “Transfers to agents directly in preferred Inbox”

   proof ↗⚑ flag

   Decagon~partialclaimed4/10

   Decagon's AOPs let teams define agent behavior in natural language (like SOPs), which could include escalation logic, and Watchtower can flag sentiment/compliance signals, but no evidence explicitly describes configuring handoff triggers by topic, sentiment, customer tier, or explicit request, nor confirms reliable adherence to such rules. Simulations/testing-QA claims validating 'policies' are generic and don't specifically address handoff reliability. Missing for 10: explicit documentation of handoff/escalation configuration options, tier-based routing, and evidence of reliable handoff obedience (e.g., test results or case studies on escalation accuracy).

   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Whether you're monitoring for compliance risks, negative sentiment, or upsell signals, Watchtower reviews every interaction against your cus…”
   • [claimed-docs] “Non-technical teams can architect and iterate on agent logic with Agent Operating Procedures (AOPs) and Duet, while technical teams retain f…”
   • [claimed-docs] “Simulations, Decagon's integrated testing suite, help teams validate agent behavior across channels before deploying to production and with …”

   proof ↗⚑ flag

Hallucination

1. ai-native userGuardrails stop the agent from inventing policies, prices, or promises — off-knowledge questions get a safe decline, not a guess

   weight 3 · round to Fin
   Fin~partialclaimed6/10

   Docs show Fin is 'trained on your knowledge, data, policies, and tone' and grounded in the knowledge base (docs-46/60/49), with configurable 'escalation guidance and rules' (docs-25) and pre-launch testing across scenarios (docs-45/59), which together imply guardrails against off-knowledge answers and hand-off rather than free invention. However there is no explicit documentation of a 'safe decline' behavior or hallucination-refusal mechanism, and no independent/hands-on evidence confirming Fin actually declines rather than guesses on off-knowledge questions. Missing for 10: explicit safe-decline/refusal documentation, third-party or hands-on verification that Fin avoids inventing policies/prices, and quantified guardrail testing results.

   • [claimed-docs] “Train Fin on your knowledge, data, policies, and tone of voice.”
   • [claimed-docs] “Train Fin on your knowledge, data, policies, and tone of voice. Fin will use everything it knows to deliver consistent, accurate answers”
   • [claimed-docs] “Manage Fin AI Agent's escalation guidance and rules”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation... …”
   • [claimed-docs] “grounded in your knowledge base, and trained to apply your specific policies on every call.”
   • [claimed-docs] “give you the tools to safely deploy AI within your own boundaries.”
   • [claimed-docs] “We meet the world's leading compliance standards—and give you the tools to safely deploy AI within your own boundaries.”

   proof ↗⚑ flag

   Decagon~partialclaimed4/10

   Decagon provides AOPs to define agent behavior/policies in natural language and a testing suite (Simulations/Duet) that verifies agents 'respond accurately, follow policies' plus Watchtower monitoring for compliance violations, which are the building blocks for guardrails — but no evidence explicitly describes a safe-decline behavior for off-knowledge questions versus fabricating answers. missing for 10: explicit documentation or examples of the agent refusing/declining out-of-scope questions rather than hallucinating, and independent verification that guardrails actually prevent invented prices/policies in practice.

   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”
   • [claimed-docs] “Use Duet to easily generate tests covering diverse pathways that verify whether agents respond accurately, follow policies, and reflect your…”
   • [claimed-docs] “Simulations, Decagon's integrated testing suite, help teams validate agent behavior across channels before deploying to production and with …”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Whether you're monitoring for compliance risks, negative sentiment, or upsell signals, Watchtower reviews every interaction against your cus…”

   proof ↗⚑ flag

Supervision

1. support ops leadLaunch in a supervised mode where the agent drafts replies for human approval before anything reaches a customer

   weight 2 · round drawn
   Fin—none0/10

   Evidence shows Fin's guardrails center on escalation rules, hand-off to human agents, and pre-launch testing/preview (docs-25, docs-29, docs-31, docs-45), but nothing describes a supervised/draft-for-approval mode where Fin composes replies that a human must approve before they reach the customer on every interaction. Community commentary (comm-3) even argues for more deterministic, human-controlled workflows, but that's not evidence Fin ships this specific approval-gate mode.

   • [claimed-docs] “Manage Fin AI Agent's escalation guidance and rules”
   • [claimed-docs] “How to preview and test Fin”
   • [claimed-docs] “Hand-off with JavaScript”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation”
   • [community] “A commenter with a decade of customer-service AI/NLP experience argued 99% of support tasks (order status, returns, troubleshooting) are ful…”

   proof ↗⚑ flag

   Decagon—none0/10

   Decagon's evidence covers testing/simulation (Simulations, Duet), experiments, guardrails, and monitoring (Watchtower), but nothing describes a supervised/human-in-the-loop 'draft-for-approval' mode where a human must approve each reply before it reaches a customer. Testing/QA and experiments happen pre-production or on traffic splits, not as a live approval queue for individual replies.

   ⚑ flag

Topic controls

1. support ops leadI mark topics as human-only — legal threats, cancellations, security — and the agent never freelances on them

   weight 2 · round to Fin
   Fin~partialclaimed6/10

   Fin has documented first-party features for exactly this: 'Giving Fin Guidance' and 'Manage Fin AI Agent's escalation guidance and rules' let ops teams define escalation/human-handoff rules, and the Agent API includes an 'Escalate to a human' capability. However this escalate-to-human orchestration endpoint is explicitly noted as Preview-only, and there is no independent or hands-on evidence confirming the agent reliably respects human-only topic boundaries without ever freelancing. Missing for 10: independent verification that escalation rules are strictly enforced (no freelancing), and GA (non-preview) status of escalate-to-human tooling.

   • [claimed-docs] “Giving Fin Guidance”
   • [claimed-docs] “Manage Fin AI Agent's escalation guidance and rules”
   • [claimed-docs] “The new orchestration endpoints — Discover capabilities, Ask Fin, Run a procedure, and Escalate to a human — are currently available in the …”

   proof ↗⚑ flag

   Decagon~partialclaimed5/10

   Decagon's AOPs let teams define agent behavior/escalation rules in natural language, and 'guardrails' are described as retained under technical team control, which could support marking certain topics as human-only; Watchtower can flag legal/compliance/sentiment topics for review. However, there's no explicit documentation of a dedicated 'human-only topic' or hard-escalation feature, nor any evidence (hands-on or audit) confirming the agent never engages on flagged categories. Missing for 10: explicit human-only/topic-exclusion configuration feature, proof/testing that the agent reliably refuses or escalates on those topics, and independent verification of enforcement.

   • [claimed-docs] “technical teams retain full visibility and control over guardrails, integrations, and versioning”
   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”
   • [claimed-docs] “Non-technical teams can architect and iterate on agent logic with Agent Operating Procedures (AOPs) and Duet, while technical teams retain f…”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Whether you're monitoring for compliance risks, negative sentiment, or upsell signals, Watchtower reviews every interaction against your cus…”

   proof ↗⚑ flag

Analytics

1. support leaderDashboards show resolution rate, CSAT, handoff rate, and cost per resolution — the numbers I report to my exec team

   weight 3 · round to Decagon
   Fin~partialclaimed4/10

   Fin's Insights/Monitors/AI Recommendations pages reference analyzing conversations and measuring CX, and CX Score/Trends are listed as features, implying some dashboard capability, but no evidence explicitly confirms a unified dashboard reporting resolution rate, CSAT, handoff rate, and cost per resolution together. missing for 10: explicit documentation or screenshots of a dashboard showing resolution rate, CSAT, handoff rate, and cost per resolution metrics; independent confirmation these specific KPIs are surfaced for exec reporting.

   • [claimed-docs] “Insights continuously analyzes every Fin and human conv”
   • [claimed-docs] “With Insights, Monitors New, and Recommendations, you have everything you need to understand, measure, and improve your customer experience”
   • [claimed-docs] “Operator, CX Score, AI Topics, Trends, AI Recommendations, Monitors, Custom AI Scorecards, Incident Detection”
   • [claimed-docs] “Fin resolves 76% of customer queries on average, and handles multi-step queries end to end. Your customers get fast, accurate answers. Your …”

   proof ↗⚑ flag

   Decagon~partialclaimed5/10

   Decagon's Insights & Reporting product includes customizable dashboards, visual heatmaps for key metrics, and 'Ask AI' conversational analytics for querying conversation trends, which supports general exec-reporting use cases. However, the evidence never explicitly names resolution rate, CSAT, handoff rate, or cost-per-resolution as tracked/reported metrics. missing for 10: explicit confirmation that CSAT, resolution rate, handoff rate, and cost-per-resolution are specific dashboard metrics; independent/customer corroboration of these exact KPIs being reported to execs.

   • [claimed-docs] “Ask open-ended questions like “Why are customers requesting refunds?” to instantly analyze customer conversations.”
   • [claimed-docs] “Customizable visual heatmaps... Quickly identify spikes or dips in key metrics.”
   • [claimed-docs] “Ask AI powers deeper analysis through flexible, conversational queries.”

   proof ↗⚑ flag

Insights

1. support leaderThe platform clusters conversations by topic and surfaces emerging product issues before they spike ticket volume

   weight 1 · round to Decagon
   Fin~partialclaimed6/10

   Fin's pricing page lists analytics features like 'AI Topics, Trends... Monitors... Incident Detection' and 'Insights continuously analyzes every Fin and human conversation,' indicating topic clustering and emerging-issue detection capability, but there is no deeper documentation, screenshots, or independent corroboration of how this works in practice. Missing for 10: detailed product docs on topic clustering methodology, evidence of proactive alerting before ticket-volume spikes, and independent/hands-on validation of the insights/analytics feature.

   • [claimed-docs] “Operator, CX Score, AI Topics, Trends, AI Recommendations, Monitors, Custom AI Scorecards, Incident Detection”
   • [claimed-docs] “Insights continuously analyzes every Fin and human conv”
   • [claimed-docs] “With Insights, Monitors New, and Recommendations, you have everything you need to understand, measure, and improve your customer experience”

   proof ↗⚑ flag

   Decagon~partialclaimed7/10

   Decagon's Insights & Reporting and Watchtower products support natural-language querying of conversations (e.g., 'Why are customers requesting refunds?'), heatmaps to spot spikes/dips in key metrics, and custom flagging criteria across every interaction, which together enable topic-level pattern detection and early issue surfacing. However, the evidence never explicitly describes automated topic clustering or a proactive 'before ticket volume spikes' alerting mechanism—these are inferred from adjacent features. Missing for 10: explicit description of automatic conversation clustering by topic, independent/hands-on evidence of early trend detection preventing ticket spikes, and case-study proof of this specific insight workflow in production.

   • [claimed-docs] “Ask open-ended questions like “Why are customers requesting refunds?” to instantly analyze customer conversations.”
   • [claimed-docs] “Customizable visual heatmaps... Quickly identify spikes or dips in key metrics.”
   • [claimed-docs] “Ask AI powers deeper analysis through flexible, conversational queries.”
   • [claimed-docs] “Whether you're monitoring for compliance risks, negative sentiment, or upsell signals, Watchtower reviews every interaction against your cus…”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues, so your…”

   proof ↗⚑ flag

Helpdesk

1. developerThe agent runs inside my existing helpdesk — Zendesk, Salesforce, Intercom — or standalone, syncing tickets and context both ways

   weight 3 · round to Fin
   Fin✓fullclaimed8/10

   Fin explicitly documents standalone deployment plus deep integrations with Zendesk, Salesforce, HubSpot, Freshdesk and 'any helpdesk', with two-way data sync (bring content/history over, surface data in inbox) and APIs/MCP/webhooks for programmatic ticket/context sync. Multiple dedicated integration pages (Zendesk, Salesforce, HubSpot, Freshdesk) and platform docs (Agent API, MCP, webhooks, data connectors) corroborate bidirectional context flow. missing for 10: independent/hands-on verification of the bidirectional sync working in production, and no direct evidence of Intercom-native standalone ticketing depth compared to third-party helpdesks.

   • [claimed-docs] “Fin AI Agent works seamlessly with any helpdesk, including Salesforce, HubSpot, and more.”
   • [claimed-docs] “How Fin integrates with Zendesk”
   • [claimed-docs] “How Fin integrates with Salesforce”
   • [claimed-docs] “bring your team over from any helpdesk without losing your content, your history, or a single day of support.”
   • [claimed-docs] “Surface data from Salesforce, HubSpot, and 450+ apps and integrations directly in the inbox”
   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [claimed-docs] “Learn how to use the Model Context Protocol (MCP) to enable AI agents to securely access and interact with your Intercom data whenever helpf…”
   • [claimed-docs] “Webhooks allow you to subscribe to real-time notifications of events happening in Intercom; a Contact created, an incoming Conversation rece…”
   • [claimed-docs] “It works through MCP or API Data Connectors for your business tools, integrations with Intercom or your existing helpdesk”
   • [claimed-docs] “Intercom connects to 450+ first- and third-party apps and integrations, including CRMs like Salesforce and HubSpot, dev tools like Jira, Lin…”

   proof ↗⚑ flag

   Decagon~partialclaimed4/10

   Decagon's integrations page claims no-code integrations and system-of-record data portability, and Browser Actions/MCP posts describe connecting to arbitrary systems including ones without native integrations, implying support for helpdesk platforms like Zendesk/Salesforce/Intercom, but no evidence explicitly names these tools or describes ticket-level two-way sync. Missing for 10: named connectors/case studies for Zendesk, Salesforce, or Intercom, explicit description of bidirectional ticket-context syncing, and confirmation of standalone deployment mode.

   • [claimed-docs] “no custom code required”
   • [claimed-docs] “with no custom code required”
   • [claimed-docs] “Decagon integrates with identity providers like Okta and Microsoft Entra, enabling secure access across systems without additional passwords…”
   • [claimed-docs] “Maintain customer context across sessions, complementing your systems of record with data portability and control.”
   • [claimed-docs] “Browser Actions gives your agent a way through computer use. It extends the agent you've already built so it can log in, navigate, and compl…”
   • [claimed-docs] “lets your Decagon agent access and complete tasks inside systems where there isn’t a traditional integration available”
   • [claimed-docs] “Bridging that gap requires an infrastructure layer purpose-built to curate, scope, and evaluate how tools are actually used.”

   proof ↗⚑ flag

Freshness

1. support ops leadKnowledge stays current automatically — the agent re-syncs sources on a schedule or on change, not via manual re-uploads

   weight 2 · round to Fin
   Fin~partialclaimed5/10

   Docs mention 'Sync and manage websites' and content import via API (createContentImportSource), implying re-syncing of website sources, but there is no explicit documentation of scheduled or change-triggered automatic re-syncing across all source types (docs, help centers, files) without manual re-upload. missing for 10: explicit scheduling/cron documentation for content refresh, confirmation of automatic change-detection re-sync for non-website sources (PDFs, articles), and independent/hands-on verification that sync happens without manual re-upload.

   • [claimed-docs] “Sync and manage websites”
   • [github] “await client.aiContent.createContentImportSource({ url: "https://www.example.com" });”

   proof ↗⚑ flag

   Decagon—none0/10

   The evidence pack covers knowledge gap detection and content suggestions (docs-10/18), integrations without custom code, and agent iteration via Duet, but nowhere mentions automatic re-syncing of knowledge sources on a schedule or on change — the closest analog (Suggestions) generates draft content for humans to review/publish, not an automated re-sync pipeline. Missing for 10: any mention of scheduled/triggered re-ingestion of source documents, sync frequency, or change-detection on connected knowledge bases.

   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues, so your…”
   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues”

   proof ↗⚑ flag

Gaps

1. support ops leadThe platform surfaces knowledge gaps and conflicting content that cause the agent to miss or fumble questions

   weight 1 · round to Decagon
   Fin~partialclaimed4/10

   Fin's analytics stack (Insights, Monitors, AI Topics/Trends, AI Recommendations, Incident Detection) implies some ability to flag where Fin underperforms or topics trend poorly, and 'Giving Fin Guidance'/'Adding content to Fin' suggest content curation workflows, but no evidence explicitly describes detecting knowledge gaps or conflicting source content as a distinct surfaced capability. Missing for 10: explicit documentation of a 'knowledge gap' or 'content conflict' detection feature, and any hands-on/independent confirmation that such gaps are surfaced to support-ops leads.

   • [claimed-docs] “Operator, CX Score, AI Topics, Trends, AI Recommendations, Monitors, Custom AI Scorecards, Incident Detection”
   • [claimed-docs] “Insights continuously analyzes every Fin and human conv”
   • [claimed-docs] “With Insights, Monitors New, and Recommendations, you have everything you need to understand, measure, and improve your customer experience”
   • [claimed-docs] “Giving Fin Guidance”
   • [claimed-docs] “Adding content to Fin”

   proof ↗⚑ flag

   Decagon✓fullclaimed7/10

   Decagon's Suggestions product explicitly automates detection of knowledge gaps and drafts content based on how top human agents resolved similar issues, directly addressing the story's core need, and Insights/Watchtower let ops teams query conversations to surface where the agent struggles or conflicts arise. Missing for 10: independent/hands-on verification of gap detection accuracy, and no explicit mention of surfacing 'conflicting content' across knowledge sources rather than just gaps.

   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues, so your…”
   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues”
   • [claimed-docs] “Ask open-ended questions like “Why are customers requesting refunds?” to instantly analyze customer conversations.”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Customizable visual heatmaps... Quickly identify spikes or dips in key metrics.”

   proof ↗⚑ flag

Grounding

1. ai-native userEvery answer is grounded in my own content and shows which article or source it drew from

   weight 3 · round to Fin
   Fin~partialclaimed5/10

   Fin's docs confirm it is trained/grounded on the customer's own content (help center, websites, docs, knowledge sources) via content import and knowledge retrieval APIs (e.g., docs-46/60 'train Fin on your knowledge', docs-28 'Adding content to Fin', docs-27 'Sync and manage websites', gh-1 content import API, docs-39 knowledge retrieval API), but no evidence explicitly shows that end-user answers surface or cite the specific source article they were drawn from. missing for 10: explicit documentation or UI evidence of in-answer source citation/attribution, independent/hands-on confirmation of citation behavior.

   • [claimed-docs] “Train Fin on your knowledge, data, policies, and tone of voice.”
   • [claimed-docs] “Train Fin on your knowledge, data, policies, and tone of voice. Fin will use everything it knows to deliver consistent, accurate answers”
   • [claimed-docs] “Adding content to Fin”
   • [claimed-docs] “Sync and manage websites”
   • [github] “await client.aiContent.createContentImportSource({ url: "https://www.example.com" });”
   • [claimed-docs] “The Fin API platform includes APIs for answer generation and knowledge retrieval, available as a set or independently.”
   • [claimed-docs] “grounded in your knowledge base, and trained to apply your specific policies on every call.”

   proof ↗⚑ flag

   Decagon—none0/10

   Decagon's evidence describes knowledge-gap detection, AOP generation, and content suggestions (docs-10, docs-18) but nothing shows that end-user-facing answers actually cite or display the specific article/source they were grounded in. Missing for 10: any documentation of inline citations, source attribution UI, or 'view source' feature in chat/voice/email responses.

   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues, so your…”
   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues”
   • [claimed-docs] “Flexible, on-brand chat built for conversations across web, mobile, and messaging platforms.”

   proof ↗⚑ flag

Ingestion

1. support ops leadThe agent ingests my help center, docs, past tickets, and internal wikis as knowledge sources without manual re-authoring

   weight 3 · round to Fin
   Fin✓fullclaimed8/10

   Fin explicitly supports ingesting websites/help center content (createContentImportSource, 'Sync and manage websites', 'Adding content to Fin'), integrates with helpdesks (Zendesk, Salesforce, HubSpot, Freshdesk) preserving existing content/history, and pulls in Confluence, Notion, Zendesk content into a unified knowledge source alongside data connectors for third-party tools. This covers help center, docs, past ticket/helpdesk content, and wiki tools (Confluence/Notion) without manual re-authoring. Missing for 10: no explicit mention of ingesting 'internal wikis' broadly beyond Confluence/Notion, and no independent/hands-on verification of ingestion fidelity or effort required.

   • [github] “await client.aiContent.createContentImportSource({ url: "https://www.example.com" });”
   • [claimed-docs] “Sync and manage websites”
   • [claimed-docs] “Adding content to Fin”
   • [claimed-docs] “Bring content from your help center, Zendesk, Confluence, Notion, and more into one place, so Fin, Copilot, and your team always work from t…”
   • [claimed-docs] “bring your team over from any helpdesk without losing your content, your history, or a single day of support.”
   • [claimed-docs] “How Fin integrates with Zendesk”
   • [claimed-docs] “How Fin integrates with Salesforce”
   • [claimed-docs] “Your team stays on Freshdesk.”

   proof ↗⚑ flag

   Decagon~partialclaimed4/10

   Decagon's integrations page claims 'no custom code required' connections (docs-19/26) and its Suggestions product analyzes help-center content and past agent resolutions to fill knowledge gaps (docs-10/18), implying some ingestion of existing content, but there is no explicit documentation naming help center, internal wikis, docs, and past tickets as ingestible knowledge sources without re-authoring. missing for 10: explicit list/documentation of supported knowledge source connectors (help center, wikis, ticket systems), evidence of automatic re-indexing without manual content rewriting, and independent confirmation of successful multi-source ingestion.

   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues, so your…”
   • [claimed-docs] “It automatically detects knowledge gaps and generates content drafts based on how your best human agents resolved those same issues”
   • [claimed-docs] “no custom code required”
   • [claimed-docs] “with no custom code required”

   proof ↗⚑ flag

1. ai-native userDo everything through the API that I can do in the UI

   weight 2 · round to Fin
   Fin~partialprobed4/10

   Fin exposes an Agent API, Node SDK, MCP, webhooks, and data-connector APIs for programmatic interaction, but there is no evidence of full UI/API parity — no OpenAPI spec was discoverable (404s on all candidate paths) and no documentation claims that every configuration (e.g., inbox setup, workflows, Procedures, escalation rules) manageable in the UI is also manageable via API. Missing for 10: comprehensive API reference/OpenAPI spec, explicit parity claims, evidence that admin/config UI actions (not just conversational/data actions) are API-accessible.

   • [claimed-docs] “Fin can be accessed programmatically via an API. You call Fin from your o”
   • [claimed-docs] “The new orchestration endpoints — Discover capabilities, Ask Fin, Run a procedure, and Escalate to a human — are currently available in the …”
   • [github] “The Intercom TypeScript library provides convenient access to the Intercom APIs from TypeScript.”
   • [probe] “PROBE openapi: all candidate paths 404 (https://fin.ai/openapi.json, https://fin.ai/swagger.json, https://fin.ai/api/openapi.json, https://f…”
   • [claimed-docs] “The Fin API platform includes APIs for answer generation and knowledge retrieval, available as a set or independently.”

   proof ↗⚑ flag

   Decagon—none0/10

   No evidence of a public API at all — the openapi probe returned 404 across all candidate paths, and none of the docs describe an API that mirrors UI capabilities like AOP configuration, Duet, Watchtower, or experiments. Absence of evidence for this applicable capability yields none.

   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

2. ai-native userExport all of my data in open formats and leave

   weight 3 · round to Fin
   Fin~partialclaimed2/10

   The only export-related evidence is a narrow feature to 'Export a saved View exactly as you see it,' not a comprehensive data export mechanism; migration language found in the evidence is framed only around bringing data INTO Intercom from other helpdesks, not exporting all data out in open formats. Missing for 10: bulk/full account data export, documented open file formats (CSV/JSON), a migration-out or offboarding process, and any independent confirmation of successful full data extraction.

   • [claimed-docs] “Export a saved View exactly as you see it”
   • [claimed-docs] “bring your team over from any helpdesk without losing your content, your history, or a single day of support.”
   • [claimed-docs] “bring your team over from any helpdesk without losing your content, your history, or a single day of support”

   proof ↗⚑ flag

   Decagon—none0/10

   Decagon's evidence pack contains no documented data-export feature, open-format export tooling, or account-deletion/portability workflow; the only tangential mention ("data portability and control" in decagon-docs-28) is vague marketing language about maintaining conversation context, not a concrete export/exit mechanism, and the probe shows no public API/OpenAPI spec that could support programmatic data extraction.

   • [claimed-docs] “Maintain customer context across sessions, complementing your systems of record with data portability and control.”
   • [probe] “PROBE openapi: all candidate paths 404 (https://decagon.ai/openapi.json, https://decagon.ai/swagger.json, https://decagon.ai/api/openapi.jso…”

   proof ↗⚑ flag

Pricing

1. support leaderPricing is outcome-based and published — I pay per resolution with caps and controls, not an opaque enterprise quote

   weight 2 · round to Fin
   Fin~partialcommunity4/10

   A community source states Fin charges '$1 per successful customer session', which is consistent with outcome-based, per-resolution pricing, and a dedicated fin.ai/pricing page exists (cited repeatedly for feature lists). However, none of the evidence shows the actual published rate table, caps, or usage controls on that pricing page — the citations only reference features (integrations, channels, AI Agent capabilities), not price mechanics or caps. Missing for 10: first-party documentation of the per-resolution rate/caps/controls on fin.ai/pricing, and independent corroboration beyond a single HN comment.

   • [community] “Fin charges $1 per successful customer session, incentivizing it to be helpful; at the time of an earlier interview Fin was reportedly still…”
   • [claimed-docs] “Fin AI Agent works seamlessly with any helpdesk, including Salesforce, HubSpot, and more.”

   proof ↗⚑ flag

   Decagon—none0/10

   No evidence pack item mentions pricing, resolution-based billing, caps, controls, or published pricing pages — all evidence concerns product features (AOPs, Duet, Watchtower, integrations, security) rather than pricing model. Absence of evidence for this applicable axis yields 'none'. Missing for 10: published pricing page, resolution-based billing structure, caps/controls documentation, any pricing transparency claims.

   ⚑ flag

1. ai-native userChoose where my data is stored (region/residency)

   weight 2 · round drawn
   Fin—none0/10

   The evidence pack contains extensive documentation on Fin's features, integrations, and compliance messaging (e.g., 'meet the world's leading compliance standards') but no mention of data residency, region selection, or storage location controls anywhere in the docs or community evidence.

   ⚑ flag

   Decagon—none0/10

   No evidence pack item mentions data residency, region selection, or storage location controls; security page content covers JWTs and SSO but not data residency options. Missing for 10: any mention of regional data storage, residency compliance (e.g., EU/US hosting options), or customer-facing controls to choose storage location.

   ⚑ flag

2. ai-native userPrevent my data from being used to train AI models

   weight 3 · round drawn
   Fin—none0/10

   No evidence in the pack addresses data being used for AI model training or an opt-out/data-use control; general trust/compliance mentions (e.g., trust-reliability page) do not specify training data usage or opt-out mechanisms.

   ⚑ flag

   Decagon—none0/10

   No evidence in the pack addresses data usage for AI model training or an opt-out/no-training policy; the security page covers JWT tokens and SSO but not training data practices.

   ⚑ flag

3. ai-native userControl data retention and deletion

   weight 2 · round to Decagon
   Fin—none0/10

   No evidence in the pack addresses data retention policies, deletion controls, or user-facing controls over how long data is kept or when it is deleted; the docs cover integrations, channels, MCP access, and training but never retention/deletion settings.

   ⚑ flag

   Decagon~partialclaimed3/10

   Decagon only vaguely references 'data portability and control' alongside customer context retention, and short-lived JWT tokens discarded after each session, but there is no explicit documentation of configurable data retention periods, deletion APIs, or user-facing controls to purge stored customer data. missing for 10: explicit retention policy/settings, a documented deletion mechanism or API, and independent verification that deletion requests are honored.

   • [claimed-docs] “Maintain customer context across sessions, complementing your systems of record with data portability and control.”
   • [claimed-docs] “Short-lived JWT tokens give AI agents real-time access to customer systems, scoped for minimal privilege and discarded after each session.”

   proof ↗⚑ flag

4. ai-native userOpt out of telemetry and usage tracking

   weight 2 · round drawn
   Fin—none0/10

   No evidence pack items mention telemetry opt-out, usage tracking controls, or privacy settings for Fin; only trust/compliance marketing language and unrelated docs are present.

   ⚑ flag

   Decagon—none0/10

   No evidence in the pack addresses telemetry/usage-tracking opt-out controls; the security and product pages cover access control, JWT tokens, and SSO but never mention analytics/telemetry opt-out settings for end users. Missing for 10: any documentation of a telemetry toggle, data-collection opt-out mechanism, or privacy settings page.

   ⚑ flag

Personalization

1. support leaderAnswers use the customer's live data — plan, order status, account history — not just generic help articles

   weight 2 · round to Fin
   Fin✓fullprobed7/10

   Docs show Fin connects to live business data via data connector templates (Stripe, Shopify, Statuspage), MCP/API data connectors, webhooks, and CRM integrations (Salesforce, HubSpot, Zendesk) that surface data like plan/order status directly in the inbox, plus Procedures that let Fin execute multi-step logic against third-party systems to update/retrieve real account data rather than just canned articles. Missing for 10: independent/hands-on verification that live data grounding actually improves accuracy in practice, and no case study quantifying resolution based specifically on live account/order data vs. help-article content.

   • [claimed-docs] “Data connector templates are now generally available, with ready-to-install templates across Stripe, Shopify and Statuspage.”
   • [claimed-docs] “Track data connector health over time”
   • [claimed-docs] “Takes action to update external systems”
   • [claimed-docs] “See everything in one view in the inbox — pull in data from tools like Shopify and Stripe alongside the conversation, so agents never have t…”
   • [claimed-docs] “Surface data from Salesforce, HubSpot, and 450+ apps and integrations directly in the inbox”
   • [claimed-docs] “Procedures let you train Fin to handle queries with multiple steps, business logic and third-party systems from start to finish.”
   • [claimed-docs] “MCP is a protocol that enables AI tools and applications to securely access and interact with your Intercom data whenever helpful.”
   • [probe] “official MCP server documented at https://developers.intercom.com/docs/guides/mcp”

   proof ↗⚑ flag

   Decagon~partialclaimed6/10

   Decagon's docs describe real-time, scoped access to customer systems via short-lived JWTs, MCP-based tool integrations, and Browser Actions that let agents log into and pull data from any system (including those without native integrations), which supports live-data-driven answers rather than static help content. However, there is no concrete example or case study showing an actual answer surfacing a customer's plan, order status, or account history — the evidence is architectural/capability-level rather than demonstrated output. Missing for 10: a documented example or case study showing live plan/order/account data appearing in an actual customer-facing answer, and independent verification that this works in production.

   • [claimed-docs] “Short-lived JWT tokens give AI agents real-time access to customer systems, scoped for minimal privilege and discarded after each session.”
   • [claimed-docs] “Bridging that gap requires an infrastructure layer purpose-built to curate, scope, and evaluate how tools are actually used.”
   • [claimed-docs] “Browser Actions gives your agent a way through computer use. It extends the agent you've already built so it can log in, navigate, and compl…”
   • [claimed-docs] “lets your Decagon agent access and complete tasks inside systems where there isn’t a traditional integration available”
   • [claimed-docs] “no custom code required”
   • [claimed-docs] “Decagon integrates with identity providers like Okta and Microsoft Entra, enabling secure access across systems without additional passwords…”

   proof ↗⚑ flag

Reasoning

1. support leaderThe agent asks clarifying questions and works through multi-step troubleshooting instead of dumping one canned answer

   weight 2 · round to Fin
   Fin~partialcommunity6/10

   Fin's docs describe 'Procedures' that train it to handle multi-step queries with business logic across systems, and marketing claims it 'handles multi-step queries end to end,' supporting structured troubleshooting rather than single canned replies. However, no evidence explicitly documents Fin asking clarifying questions mid-conversation, and one HN commenter argues open-ended agentic reasoning is unnecessary versus deterministic workflows, adding some uncertainty. missing for 10: explicit documentation/example of clarifying-question behavior, independent hands-on proof of multi-step troubleshooting quality.

   • [claimed-docs] “Procedures let you train Fin to handle queries with multiple steps, business logic and third-party systems from start to finish.”
   • [claimed-docs] “Fin resolves 76% of customer queries on average, and handles multi-step queries end to end. Your customers get fast, accurate answers. Your …”
   • [claimed-docs] “Giving Fin Guidance”
   • [community] “A commenter with a decade of customer-service AI/NLP experience argued 99% of support tasks (order status, returns, troubleshooting) are ful…”

   proof ↗⚑ flag

   Decagon~partialclaimed5/10

   Decagon docs describe AOPs that define multi-step agent behavior like human SOPs and mention agents carrying context across sessions and taking multi-step actions (browser actions, integrations), which implies some troubleshooting flow, but there is no direct documentation or example showing the agent asking clarifying questions or walking through iterative multi-step troubleshooting dialogues. missing for 10: concrete example/transcript of clarifying-question behavior, documentation explicitly describing multi-turn troubleshooting logic, independent/hands-on evidence confirming this behavior in practice.

   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”
   • [claimed-docs] “Carry conversation history across sessions to pick up where you left off and make proactive recommendations based on customer signals.”
   • [claimed-docs] “lets your Decagon agent access and complete tasks inside systems where there isn’t a traditional integration available”

   proof ↗⚑ flag

Resolution

1. support leaderThe agent fully resolves a meaningful share of conversations end-to-end — measured as resolutions, not mere deflections or bounces

   weight 3 · round to Fin
   Fin✓fullcommunity7/10

   Fin explicitly claims a resolution metric ("Fin resolves 76% of customer queries on average, and handles multi-step queries end to end") across multiple integration pages, plus dedicated testing/evaluation tooling ("evaluate every live conversation") and analytics (Insights, Monitors) framed around resolution outcomes rather than deflection. Community commentary raises philosophical skepticism about autonomous agents vs. deterministic workflows, but does not concretely contradict the stated resolution rate. Missing for 10: independent/third-party audit of the 76% figure, and clear definition distinguishing 'resolution' from deflection/bounce in the metric methodology.

   • [claimed-docs] “Fin resolves 76% of customer queries on average, and handles multi-step queries end to end. Your customers get fast, accurate answers. Your …”
   • [claimed-docs] “Perfect customer experiences no longer come with a migration project. Fin resolves 76% of customer queries on average... Your team stays on …”
   • [claimed-docs] “Perfect customer experiences no longer come with a migration project... Your team stays on HubSpot.”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation... …”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation”
   • [claimed-docs] “With Insights, Monitors New, and Recommendations, you have everything you need to understand, measure, and improve your customer experience”
   • [community] “A commenter with a decade of customer-service AI/NLP experience argued 99% of support tasks (order status, returns, troubleshooting) are ful…”

   proof ↗⚑ flag

   Decagon~partialclaimed4/10

   Decagon's docs emphasize automation infrastructure (AOPs, Duet, Browser Actions, integrations) and analytics (Watchtower, Insights) but provide no quantified resolution-rate metrics, no third-party benchmark, and the one customer quote (Duolingo) speaks to maintenance effort, not resolution outcomes. Missing for 10: independent or customer-reported resolution-rate figures, a clear definition/measurement of 'resolution' vs deflection, and case studies quantifying end-to-end conversation completion rather than agent capability lists.

   • [claimed-docs] “Browser Actions gives your agent a way through computer use. It extends the agent you've already built so it can log in, navigate, and compl…”
   • [claimed-docs] “lets your Decagon agent access and complete tasks inside systems where there isn’t a traditional integration available”
   • [claimed-docs] “With the previous vendor, at least half my week was dedicated to maintaining their system. With Decagon, it’s been a night-and-day differenc…”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Whether you're monitoring for compliance risks, negative sentiment, or upsell signals, Watchtower reviews every interaction against your cus…”

   proof ↗⚑ flag

Voice

1. support leaderI control the agent's tone and brand voice, and it stays consistent across topics and languages

   weight 1 · round to Decagon
   Fin~partialclaimed6/10

   Docs support tone/brand control ("Customizable tone and answer length", "Train Fin on your knowledge, data, policies, and tone of voice") and multi-language support is documented ("Use Fin AI Agent in multiple languages"), plus multi-channel consistency claims (email, Slack, live chat, voice) suggest tone carries across surfaces. However, there is no independent evidence or hands-on validation that tone/brand voice actually stays consistent across topics and languages in practice — only vendor marketing claims. Missing for 10: independent/third-party validation of tone consistency, concrete examples of cross-language tone fidelity, and evidence of consistency across many topics rather than just a training feature description.

   • [claimed-docs] “Customizable tone and answer length”
   • [claimed-docs] “Train Fin on your knowledge, data, policies, and tone of voice.”
   • [claimed-docs] “Train Fin on your knowledge, data, policies, and tone of voice. Fin will use everything it knows to deliver consistent, accurate answers”
   • [claimed-docs] “Use Fin AI Agent in multiple languages”
   • [claimed-docs] “Fin for messenger brings the highest-performing AI Agent for customer service to live chat, delivering exceptional, on-brand support”
   • [claimed-docs] “Fin delivers unrivaled performance across every channel. Wherever customers reach out, Fin provides fast, accurate, personalized support”

   proof ↗⚑ flag

   Decagon~partialclaimed7/10

   Decagon's docs show brand-voice consistency built into multiple surfaces: chat, voice (multilingual), and email are all described as 'on-brand', and AOPs let support leaders define agent behavior/tone in natural language like SOPs (docs-13,14,15,16). Testing/QA and Experiments explicitly validate tone/brand across pathways and let leaders 'refine tone' safely before and after deployment (docs-11,29,32). Missing for 10: independent/hands-on evidence that tone actually stays consistent across many topics and languages in production, and no explicit mention of cross-language consistency for chat/email (only voice is called multilingual).

   • [claimed-docs] “Flexible, on-brand chat built for conversations across web, mobile, and messaging platforms.”
   • [claimed-docs] “Voice AI agents built for natural, multilingual conversations and customized to your brand.”
   • [claimed-docs] “Intelligent email that understands context, stays on brand, and handles complex customer threads.”
   • [claimed-docs] “you can define agent behavior in natural language, the same way you train human agents with SOPs”
   • [claimed-docs] “Use Duet to easily generate tests covering diverse pathways that verify whether agents respond accurately, follow policies, and reflect your…”
   • [claimed-docs] “Simulations, Decagon's integrated testing suite, help teams validate agent behavior across channels before deploying to production and with …”
   • [claimed-docs] “Whether you’re refining tone, adjusting refund logic, or trialing a new onboarding experience, experimentation makes it safe to test, learn,…”

   proof ↗⚑ flag

Qa

1. support ops leadAI conversations get ongoing QA — scored samples, flagged failures, and a review loop that feeds fixes back into the agent

   weight 1 · round to Decagon
   Fin✓fullclaimed7/10

   Fin ships explicit QA tooling: fin.ai/testing describes testing across thousands of scenarios before launch and evaluating every live conversation, fin.ai/analyze describes Insights, Monitors, and Recommendations continuously analyzing every conversation, and pricing docs list Custom AI Scorecards and Incident Detection alongside guidance and procedure-editing docs that let fixes feed back into the agent behavior. missing for 10: independent or hands-on verification of the scoring and review loop, and detail on how flagged failures are routed to specific fixes.

   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation... …”
   • [claimed-docs] “Insights continuously analyzes every Fin and human conv”
   • [claimed-docs] “With Insights, Monitors New, and Recommendations, you have everything you need to understand, measure, and improve your customer experience”
   • [claimed-docs] “Operator, CX Score, AI Topics, Trends, AI Recommendations, Monitors, Custom AI Scorecards, Incident Detection”
   • [claimed-docs] “Giving Fin Guidance”
   • [claimed-docs] “Building Fin Procedures”

   proof ↗⚑ flag

   Decagon✓fullclaimed8/10

   Decagon documents a dedicated Testing & QA suite (Simulations) that validates agent behavior pre-deploy and on updates, plus Watchtower which reviews every live interaction against custom flagging criteria (compliance, sentiment, frustration) and surfaces failures, and Duet which auto-generates tests/AOPs and iterates on the agent based on conversation gaps — together covering scored/flagged QA and a feedback loop into agent fixes. missing for 10: no independent/hands-on corroboration of scoring accuracy or the closed-loop fix cycle, and no explicit description of 'sampling' methodology for QA review.

   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Use Duet to easily generate tests covering diverse pathways that verify whether agents respond accurately, follow policies, and reflect your…”
   • [claimed-docs] “Define what matters using natural language. From "mentions of frustration" to "violations of data privacy policy," Watchtower understands yo…”
   • [claimed-docs] “Simulations, Decagon's integrated testing suite, help teams validate agent behavior across channels before deploying to production and with …”
   • [claimed-docs] “As conversations reveal gaps and new opportunities, Duet automatically iterates on your agent, from the first launch through every improveme…”
   • [claimed-docs] “Whether you're monitoring for compliance risks, negative sentiment, or upsell signals, Watchtower reviews every interaction against your cus…”

   proof ↗⚑ flag

Simulation

1. support ops leadI test the agent against historical tickets or simulated conversations before it faces real customers

   weight 2 · round to Decagon
   Fin✓fullclaimed7/10

   Fin has a dedicated testing capability documented at fin.ai/testing ("test it across thousands of scenarios before anything goes live") and a specific help article "How to preview and test Fin," directly matching the support-ops need to validate the agent before it faces customers. Missing for 10: independent/hands-on corroboration of the testing workflow and explicit mention of importing historical ticket transcripts as test scenarios rather than only generic 'scenarios'.

   • [claimed-docs] “How to preview and test Fin”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation”
   • [claimed-docs] “you test it across thousands of scenarios before anything goes live, roll out changes with control, and evaluate every live conversation... …”
   • [claimed-docs] “Train Fin on your Procedures, knowledge, and policies, test performance before launch, deploy across every channel, then analyze and improve…”

   proof ↗⚑ flag

   Decagon✓fullclaimed8/10

   Decagon's Simulations product is explicitly described as an integrated testing suite that validates agent behavior across channels before production deployment and on every update, and Duet can generate diverse test pathways to check accuracy, policy adherence, and brand voice. Missing for 10: no explicit mention of testing against historical ticket logs specifically, and no independent/hands-on evidence corroborating the testing suite's effectiveness beyond vendor docs.

   • [claimed-docs] “Simulations, Decagon's integrated testing suite, help teams validate agent behavior across channels before deploying to production and with …”
   • [claimed-docs] “Use Duet to easily generate tests covering diverse pathways that verify whether agents respond accurately, follow policies, and reflect your…”
   • [claimed-docs] “Analyze past customer interactions to automatically generate Agent Operating Procedures (AOPs) rooted in actual customer needs.”

   proof ↗⚑ flag