Decagon vs Lorikeet

Actions

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

   weight 3 · round to Lorikeet
   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

   Lorikeet✓fullclaimed7/10

   Docs explicitly state the agent 'takes action through your APIs and MCP servers' (docs-28), give concrete examples like refund workflows and order-status tool tests (docs-19, docs-24, docs-43), and describe workflow-scoped tool access, server-side identity validation, and hard execution caps enforced in code rather than prompts (docs-34). This directly matches the story of scoped, real-world API actions like refunds/order changes/subscriptions. Missing for 10: independent/hands-on verification of the auth-scoping mechanism, and more granular detail on how 'per action' scopes are configured/enforced beyond high-level guardrails language.

   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”
   • [claimed-docs] “Customer isolation, server-side identity validation, workflow-scoped tool access and hard execution caps are enforced in code rather than pr…”
   • [claimed-docs] “Test the 'get-order-status' tool with order ID 98765 and check if the response matches what we expect.”
   • [claimed-docs] “/lorikeet:create-simulations build simulations for the refund workflow”
   • [claimed-docs] “Automate disputes, loan inquiries, payments and compliance cases across every customer channel”
   • [claimed-docs] “Connect Lorikeet in seconds to your ticketing system, knowledge base, and internal tools to seamlessly ingest data and take action for your …”

   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 drawn
   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

   Lorikeet✓fullclaimed7/10

   Lorikeet explicitly supports training agents on SOPs (docs-25) and building workflows with 'natural-language and deterministic workflows' plus 'pockets of determinism for regulated steps' (docs-28, docs-32), with workflow-scoped tool access and hard execution caps enforced in code (docs-34), directly matching the deterministic-branching SOP story. Missing for 10: independent/hands-on validation of branching logic in practice and more detail on how branching conditions are authored beyond natural-language workflow builder claims.

   • [claimed-docs] “train the agent on your business context, brand guidelines, help docs and standard operating procedures”
   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”
   • [claimed-docs] “Lorikeet coordinates a team of specialist agents, with pockets of determinism for regulated steps, to handle multi-party, multi-system workf…”
   • [claimed-docs] “Customer isolation, server-side identity validation, workflow-scoped tool access and hard execution caps are enforced in code rather than pr…”
   • [claimed-docs] “Build workflows - create and iterate on workflows using natural language”
   • [claimed-docs] “Build, edit, and deploy workflows using natural language”

   proof ↗⚑ flag

Agent access

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

   weight 2 · round to Lorikeet
   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

   Lorikeet✓fullprobed9/10

   Direct probe evidence confirms Lorikeet serves an llms.txt file at docs.lorikeetcx.ai/llms.txt returning HTTP 200 with structured agent-oriented reference links, and documentation is further organized around MCP/agent access. Missing for 10: no independent third-party report of an agent successfully consuming this llms.txt in practice.

   • [probe] “PROBE llms.txt: HTTP 200 at https://docs.lorikeetcx.ai/llms.txt # Reference - [Lorikeet MCP Server](https://docs.lorikeetcx.ai/mcp/mcp-serv…”
   • [probe] “official MCP server documented at https://docs.lorikeetcx.ai/mcp/mcp-server”
   • [claimed-docs] “Connect to your Lorikeet account from ChatGPT, Claude, Claude Code, and MintMCP using the Model Context Protocol.”

   proof ↗⚑ flag

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

   weight 3 · round to Lorikeet
   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

   Lorikeet~partialclaimed5/10

   Lorikeet's docs state the agent 'takes action through your APIs and MCP servers inside natural-language and deterministic workflows' (lorikeet-docs-28), indicating it can consume external MCP servers as tool sources, but the bulk of the MCP evidence pack actually describes the reverse direction — Lorikeet exposing its own MCP server for external clients like Claude/ChatGPT to connect to (lorikeet-docs-1, lorikeet-docs-46, lorikeet-probe-1/2). Missing for 10: dedicated documentation on how a user configures/adds third-party MCP servers into Lorikeet, a list of supported MCP integrations, and independent confirmation of this client-side tool-use capability.

   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”
   • [claimed-docs] “Connect Lorikeet in seconds to your ticketing system, knowledge base, and internal tools to seamlessly ingest data and take action for your …”

   proof ↗⚑ flag

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

   weight 3 · round to Lorikeet
   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

   Lorikeet✓fullprobed7/10

   Lorikeet publishes an official MCP server (docs.lorikeetcx.ai/mcp/mcp-server) that lets Claude, ChatGPT, Claude Code, Codex, and MintMCP connect directly to a Lorikeet account, with documented capabilities like diagnosing tickets, auditing knowledge bases, building workflows, testing tools, and running simulations via MCP. This is confirmed live (HTTP 200) via probe evidence, not just marketing copy. Missing for 10: independent/hands-on third-party verification of the MCP server working end-to-end, and no community reports corroborating reliability.

   • [claimed-docs] “The Lorikeet MCP server lets you interact with your Lorikeet account directly from Claude Code, Claude.ai, ChatGPT, and Codex.”
   • [claimed-docs] “Diagnose tickets - trace workflow execution and identify root causes”
   • [claimed-docs] “Audit knowledge bases - analyze articles at scale, find gaps, and spot quality issues”
   • [claimed-docs] “Build workflows - create and iterate on workflows using natural language”
   • [claimed-docs] “Test tools - run and validate tool configurations directly from your AI assistant”
   • [claimed-docs] “Connect to your Lorikeet account from ChatGPT, Claude, Claude Code, and MintMCP using the Model Context Protocol.”
   • [probe] “PROBE llms.txt: HTTP 200 at https://docs.lorikeetcx.ai/llms.txt # Reference - [Lorikeet MCP Server](https://docs.lorikeetcx.ai/mcp/mcp-serv…”
   • [probe] “official MCP server documented at https://docs.lorikeetcx.ai/mcp/mcp-server”

   proof ↗⚑ flag

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

   weight 3 · round to Lorikeet
   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

   Lorikeet✓fullprobed8/10

   Lorikeet publishes a documented MCP server (docs.lorikeetcx.ai/mcp/mcp-server) that lets AI-native users drive the product directly from Claude, ChatGPT, Codex, and Claude Code — diagnosing tickets, auditing knowledge bases, building workflows, testing tools, and running simulations, all documented with concrete examples and even slash-command skills. This is a genuine documented programmatic interface built for AI agents, not just human UI docs. Missing for 10: no separate traditional REST/GraphQL API reference beyond MCP, and no independent third-party corroboration of the API's reliability.

   • [claimed-docs] “The Lorikeet MCP server lets you interact with your Lorikeet account directly from Claude Code, Claude.ai, ChatGPT, and Codex.”
   • [claimed-docs] “Diagnose tickets - trace workflow execution and identify root causes”
   • [claimed-docs] “Audit knowledge bases - analyze articles at scale, find gaps, and spot quality issues”
   • [claimed-docs] “Build workflows - create and iterate on workflows using natural language”
   • [claimed-docs] “Test tools - run and validate tool configurations directly from your AI assistant”
   • [claimed-docs] “Test the 'get-order-status' tool with order ID 98765 and check if the response matches what we expect.”
   • [claimed-docs] “Explore your setup - inspect workflows, tools, and integrations”
   • [claimed-docs] “Connect to your Lorikeet account from ChatGPT, Claude, Claude Code, and MintMCP using the Model Context Protocol.”
   • [claimed-docs] “Run simulations - test workflows against different customer scenarios”
   • [probe] “PROBE llms.txt: HTTP 200 at https://docs.lorikeetcx.ai/llms.txt # Reference - [Lorikeet MCP Server](https://docs.lorikeetcx.ai/mcp/mcp-serv…”
   • [probe] “official MCP server documented at https://docs.lorikeetcx.ai/mcp/mcp-server”

   proof ↗⚑ flag

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

   weight 2 · round to Decagon
   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

   Lorikeet~partialclaimed3/10

   Lorikeet's guardrails page mentions 'workflow-scoped tool access' and 'server-side identity validation' enforced in code, suggesting some access scoping, but there is no explicit documentation of issuing or managing scoped/least-privilege API credentials or tokens for agents. Missing for 10: explicit credential/token issuance mechanism, documentation of API key scoping or permission granularity, and any user-facing controls for creating least-privilege credentials.

   • [claimed-docs] “Customer isolation, server-side identity validation, workflow-scoped tool access and hard execution caps are enforced in code rather than pr…”

   proof ↗⚑ flag

6. ai-native userBuild against official SDKs

   weight 2 · round drawn
   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

   Lorikeet—none0/10

   Evidence only documents an MCP server and integrations/APIs for connecting tools, but there is no mention of an official SDK (e.g., Python/JS client library) for developers to build against. Missing for 10: any documented official SDK, its language support, or developer-facing library docs.

   ⚑ flag

7. ai-native userSubscribe to events via webhooks

   weight 2 · round drawn
   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

   Lorikeet—none0/10

   The evidence pack covers Lorikeet's MCP server, simulations, coach, and guardrails features, but contains no mention of webhooks or event subscription mechanisms for AI-native users to receive push notifications on events.

   ⚑ flag

Agentic features

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

   weight 2 · round drawn
   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

   Lorikeet✓fullclaimed8/10

   Lorikeet's Coach and MCP-server capabilities generate AI-driven insights directly from customer data: ticket quality scoring across 100% of conversations, knowledge-base gap/quality audits, root-cause diagnosis of tickets, and analytics on resolution quality, satisfaction, and revenue impact, with Coach able to 'implement improvements... or make suggestions for you to action yourself.' This is all first-party documentation without independent hands-on corroboration. Missing for 10: independent/third-party validation of insight quality and real-world usage examples beyond vendor docs.

   • [claimed-docs] “Coach's Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “Audit your entire knowledge base for gaps, outdated articles, and quality issues”
   • [claimed-docs] “Diagnose tickets by tracing workflow execution and identifying root causes”
   • [claimed-docs] “Track resolution quality, customer satisfaction, revenue impact, and operational efficiency with industry-leading analytics.”
   • [claimed-docs] “Coach can implement improvements on your behalf, or make suggestions for you to action yourself.”
   • [claimed-docs] “Coach reviews 100% of tickets against your quality standards and turns findings into fixes, so your AI agent improves every week instead of …”
   • [claimed-docs] “Scale and optimize with conversational insights and analytics from Lorikeet Coach”

   proof ↗⚑ flag

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

   weight 2 · round to Lorikeet
   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

   Lorikeet~partialclaimed7/10

   Lorikeet's core product is an autonomous agent that resolves tickets end-to-end across channels and coordinates specialist agents for multi-step workflows, and Outbound campaigns run on scheduled cadences without human intervention, while Coach can autonomously implement improvements. However, the evidence centers on the vendor's own agent running in background rather than a user-configurable 'automation' builder with triggers/schedules exposed as a general-purpose feature. Missing for 10: explicit user-facing scheduling/trigger configuration UI or API for arbitrary automations, and independent/hands-on confirmation that these automations run unattended reliably.

   • [claimed-docs] “one agent that resolves issues end-to-end across chat, email, voice and SMS”
   • [claimed-docs] “Lorikeet coordinates a team of specialist agents, with pockets of determinism for regulated steps, to handle multi-party, multi-system workf…”
   • [claimed-docs] “campaign cadences with scheduling windows control when outreach happens”
   • [claimed-docs] “Coach can implement improvements on your behalf, or make suggestions for you to action yourself.”
   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”

   proof ↗⚑ flag

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

   weight 3 · round to Lorikeet
   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

   Lorikeet✓fullclaimed7/10

   Lorikeet's 'Coach' is a built-in AI assistant accessible directly inside the Lorikeet platform (as well as via Slack/Claude/ChatGPT/MCP) that users can delegate tasks to — diagnosing tickets, auditing knowledge bases, building workflows via natural language, running simulations, and even implementing improvements automatically on the user's behalf. This is well documented across multiple first-party pages describing concrete delegated actions (e.g. doc-48 'Coach can implement improvements on your behalf'). missing for 10: independent/hands-on verification of Coach's assistant behavior, and clearer distinction between autonomous action vs. suggestion-only mode.

   • [claimed-docs] “Coach's Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “Talk to Coach wherever you work, whether in Lorikeet, Slack, Claude, ChatGPT, or via MCP.”
   • [claimed-docs] “Diagnose tickets by tracing workflow execution and identifying root causes”
   • [claimed-docs] “Audit your entire knowledge base for gaps, outdated articles, and quality issues”
   • [claimed-docs] “Build, edit, and deploy workflows using natural language”
   • [claimed-docs] “Coach reviews 100% of tickets against your quality standards and turns findings into fixes, so your AI agent improves every week instead of …”
   • [claimed-docs] “Coach can implement improvements on your behalf, or make suggestions for you to action yourself.”

   proof ↗⚑ flag

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

   weight 2 · round to Lorikeet
   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

   Lorikeet✓fullprobed8/10

   Lorikeet ships an MCP server plus Claude/ChatGPT/Codex integrations that let users build workflows, diagnose tickets, audit knowledge bases, and run simulations using natural-language commands (e.g. 'Build workflows - create and iterate on workflows using natural language', example prompts like 'Test the get-order-status tool...'), and even exposes slash-command skills like /lorikeet:create-simulations. This is first-party documentation only, with no independent/hands-on corroboration of the NL command experience. Missing for 10: independent user reports or demos validating the natural-language MCP workflow in practice.

   • [claimed-docs] “The Lorikeet MCP server lets you interact with your Lorikeet account directly from Claude Code, Claude.ai, ChatGPT, and Codex.”
   • [claimed-docs] “Build workflows - create and iterate on workflows using natural language”
   • [claimed-docs] “Build, edit, and deploy workflows using natural language”
   • [claimed-docs] “Test the 'get-order-status' tool with order ID 98765 and check if the response matches what we expect.”
   • [claimed-docs] “/lorikeet:create-simulations build simulations for the refund workflow”
   • [claimed-docs] “Connect to your Lorikeet account from ChatGPT, Claude, Claude Code, and MintMCP using the Model Context Protocol.”
   • [probe] “official MCP server documented at https://docs.lorikeetcx.ai/mcp/mcp-server”

   proof ↗⚑ flag

Api quality

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

   weight 2 · round drawn
   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

   Lorikeet—none0/10

   No evidence of a downloadable OpenAPI spec or machine-readable API documentation; evidence only covers MCP server integration, workflows, and product features, not a formal API spec artifact.

   ⚑ flag

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

   weight 1 · round drawn
   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

   Lorikeet~partialclaimed5/10

   Evidence shows robust simulation/testing tooling (replay historical tickets, synthetic scenarios, guardrail adversarial tests) that approximates sandbox-style testing, but no explicit claim of an isolated sandbox environment distinct from production. missing for 10: explicit documentation of a dedicated sandbox/staging environment, confirmation that simulations do not touch or affect live production data/systems, and independent verification of this isolation.

   • [claimed-docs] “Generate simulations straight from your actual tickets and run them in bulk batches, so every workflow change is tested against the conversa…”
   • [claimed-docs] “Run simulation batches to test workflows against different customer scenarios”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps, then deploy on the topics …”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps”
   • [claimed-docs] “Author adversarial scenarios such as false authority claims, mid-conversation goal switches and prompt injection attempts, then run them as …”
   • [claimed-docs] “Run simulations - test workflows against different customer scenarios”

   proof ↗⚑ flag

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

   weight 2 · round drawn
   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

   Lorikeet—none0/10

   No evidence pack item mentions API versioning, version numbers, or a deprecation policy for Lorikeet's APIs or MCP server; the docs discuss features and integrations but not API lifecycle/versioning commitments. Missing for 10: any documented API version scheme, changelog of breaking changes, or stated deprecation/support timeline.

   ⚑ flag

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

   weight 2 · round to Lorikeet
   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

   Lorikeet✓fullclaimed7/10

   Lorikeet documents multiple bulk operations available to AI-native users via its MCP server and product surfaces: running simulations in bulk batches across hundreds of scenarios, auditing entire knowledge bases at scale, and reviewing 100% of conversations for quality (not manual spot checks). These are explicitly framed as batch/bulk actions accessible through natural-language or MCP-driven workflows. Missing for 10: independent/hands-on verification of bulk operation scale and performance, and no explicit example of bulk edits/updates to many tickets or records simultaneously (only bulk testing/auditing/review are documented).

   • [claimed-docs] “Generate simulations straight from your actual tickets and run them in bulk batches, so every workflow change is tested against the conversa…”
   • [claimed-docs] “Side-by-side batch comparisons show how a workflow edit changed outcomes across hundreds of scenarios, with per-conversation drill-downs whe…”
   • [claimed-docs] “Audit knowledge bases - analyze articles at scale, find gaps, and spot quality issues”
   • [claimed-docs] “Audit your entire knowledge base for gaps, outdated articles, and quality issues”
   • [claimed-docs] “Coach's Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “Coach’s Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps, then deploy on the topics …”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps”
   • [claimed-docs] “Side-by-side batch comparisons show how a workflow edit changed outcomes across hundreds of scenarios, with per-conversation drill-downs”

   proof ↗⚑ flag

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

   weight 3 · round to Lorikeet
   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

   Lorikeet~partialclaimed6/10

   Lorikeet's workflows and guardrails encode conditional, event-triggered actions (e.g., prompt-injection detection triggers block/rewrite/escalate, bad QA scores trigger refunds, campaign cadences control scheduled outreach, escalation triggers hand off to humans), and workflows can be built/edited via natural language including deterministic steps. However this is more built-in platform logic than a general-purpose rule-definition interface for arbitrary custom events an AI-native user could freely wire up. Missing for 10: a documented general rules/automation engine or API letting users define arbitrary trigger-condition-action rules beyond the platform's fixed guardrail/workflow/outbound features, and independent confirmation of this working in practice.

   • [claimed-docs] “Customer isolation, server-side identity validation, workflow-scoped tool access and hard execution caps are enforced in code rather than pr…”
   • [claimed-docs] “Incoming messages pass a prompt-injection classifier and bad-actor checks, while response guardrails screen what the agent says for groundin…”
   • [claimed-docs] “When Coach gives a conversation a bad score, we refund the AI portion of that interaction.”
   • [claimed-docs] “campaign cadences with scheduling windows control when outreach happens”
   • [claimed-docs] “When a case requires a specialist or regulatory review, it escalates with full interaction history so your team picks up mid-conversation.”
   • [claimed-docs] “Lorikeet coordinates a team of specialist agents, with pockets of determinism for regulated steps, to handle multi-party, multi-system workf…”
   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”

   proof ↗⚑ flag

3. ai-native userSchedule recurring jobs or workflows

   weight 2 · round to Lorikeet
   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

   Lorikeet~partialclaimed3/10

   The only relevant evidence is outbound campaign cadences with 'scheduling windows' controlling when outreach happens, which implies some recurring/scheduled automation but is narrowly scoped to outbound messaging rather than general recurring jobs or workflow runs. Missing for 10: explicit cron-like or recurring workflow scheduling for MCP-driven tasks (simulations, audits, diagnostics), documentation of scheduling frequency/config options, and any independent confirmation of recurring job execution.

   • [claimed-docs] “campaign cadences with scheduling windows control when outreach happens”

   proof ↗⚑ flag

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

   weight 1 · round drawn
   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

   Lorikeet~partialclaimed4/10

   Lorikeet's docs show workflow building/iteration via natural language and strong review tooling (simulations, batch comparisons showing how an edit changed outcomes across scenarios), which covers the 'review' part of the story. However, there is no explicit mention of a version history or a rollback/undo mechanism for automations. missing for 10: explicit versioning/change-history feature, explicit rollback/undo capability, evidence of restoring a prior workflow state.

   • [claimed-docs] “Build workflows - create and iterate on workflows using natural language”
   • [claimed-docs] “Build, edit, and deploy workflows using natural language”
   • [claimed-docs] “Side-by-side batch comparisons show how a workflow edit changed outcomes across hundreds of scenarios, with per-conversation drill-downs whe…”
   • [claimed-docs] “Side-by-side batch comparisons show how a workflow edit changed outcomes across hundreds of scenarios, with per-conversation drill-downs”
   • [claimed-docs] “Generate simulations straight from your actual tickets and run them in bulk batches, so every workflow change is tested against the conversa…”

   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 Lorikeet
   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

   Lorikeet~partialclaimed6/10

   Lorikeet documents one agent resolving issues across chat, email, voice, SMS and WhatsApp (docs-31, docs-49), and mentions Slack as a place to interact with Coach (docs-10), but this is Coach access, not evidence that the customer-facing agent itself operates in Slack or social channels. missing for 10: explicit confirmation that the customer-facing support agent (not just Coach) handles Slack and social media channels, and independent/hands-on validation of omnichannel deployment.

   • [claimed-docs] “one agent that resolves issues end-to-end across chat, email, voice and SMS”
   • [claimed-docs] “resolve customer problems end-to-end across their lifecycle - via phone, SMS, chat, email, and WhatsApp”
   • [claimed-docs] “Talk to Coach wherever you work, whether in Lorikeet, Slack, Claude, ChatGPT, or via MCP.”
   • [claimed-docs] “When the AI concierge can't resolve a conversation, a human agent steps in directly inside the same platform”

   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 Decagon
   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

   Lorikeet—none0/10

   No evidence in the pack addresses multilingual support or the ability to serve customers in languages other than English despite an English-only knowledge base; nothing about translation, language detection, or multilingual training is mentioned.

   ⚑ 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 Decagon
   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

   Lorikeet~partialclaimed4/10

   Lorikeet's marketing claims 'one agent that resolves issues end-to-end across chat, email, voice and SMS' and lists phone/voice among supported channels, implying the same agent handles calls. However, there is no detail on speech-in/speech-out mechanics, telephony integration, or evidence that voice interactions carry the same knowledge/actions/guardrails as chat beyond a generic channel list. missing for 10: specifics on speech recognition/TTS, call-handling architecture, latency/quality benchmarks, and confirmation that voice shares the same knowledge base and action set as chat.

   • [claimed-docs] “one agent that resolves issues end-to-end across chat, email, voice and SMS”
   • [claimed-docs] “resolve customer problems end-to-end across their lifecycle - via phone, SMS, chat, email, and WhatsApp”

   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 Lorikeet
   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

   Lorikeet~partialclaimed6/10

   Lorikeet documents human handoff happening 'inside the same platform' when AI can't resolve, and specifically claims escalations carry 'full interaction history so your team picks up mid-conversation' (financial-services vertical) — directly supporting no-repeat handoff. However, evidence doesn't explicitly confirm a generated summary or structured 'collected details' package accompanying every escalation across all verticals, only interaction history. Missing for 10: explicit documentation of an auto-generated conversation summary at escalation, structured collected-details extraction, and independent/customer verification that customers never repeat themselves in practice.

   • [claimed-docs] “When the AI concierge can't resolve a conversation, a human agent steps in directly inside the same platform”
   • [claimed-docs] “When a case requires a specialist or regulatory review, it escalates with full interaction history so your team picks up mid-conversation.”
   • [claimed-docs] “Enforce required disclosures and data handling standards through built-in guardrails that balance conversational flexibility with structured…”

   proof ↗⚑ 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 Lorikeet
   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

   Lorikeet~partialclaimed5/10

   Docs show human handoff when the AI can't resolve a ticket, escalation for regulatory/specialist cases with full context, guardrails that can 'escalate' on policy violations, and deployment scoped to trained topics — supporting topic-based and inability/explicit-need escalation. However there is no explicit mention of configuring handoff by sentiment or customer tier, nor independent verification that these rules are 'reliably obeyed'. Missing for 10: explicit sentiment-based trigger config, explicit customer-tier-based trigger config, and independent/hands-on evidence of reliability.

   • [claimed-docs] “When the AI concierge can't resolve a conversation, a human agent steps in directly inside the same platform”
   • [claimed-docs] “deploy on the topics the agent is trained for and leave the rest with your team”
   • [claimed-docs] “Incoming messages pass a prompt-injection classifier and bad-actor checks, while response guardrails screen what the agent says for groundin…”
   • [claimed-docs] “When a case requires a specialist or regulatory review, it escalates with full interaction history so your team picks up mid-conversation.”
   • [claimed-docs] “Outbound runs on recorded consent that is enforced when messages are sent, with opt-out handling and escalation paths for sensitive situatio…”

   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 Lorikeet
   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

   Lorikeet✓fullclaimed8/10

   Lorikeet documents explicit guardrails that enforce grounding and policy compliance in code rather than prompts, including response screening for grounding/policy with block/rewrite/escalate actions, and adversarial simulation testing for false authority claims and prompt injection to validate safe-decline behavior before deployment. This directly targets stopping invented policies/prices/promises via server-side enforcement rather than relying on model honesty. Missing for 10: independent/third-party verification or a concrete hands-on example showing an off-knowledge question actually triggering a safe decline rather than a hallucinated answer.

   • [claimed-docs] “Customer isolation, server-side identity validation, workflow-scoped tool access and hard execution caps are enforced in code rather than pr…”
   • [claimed-docs] “Incoming messages pass a prompt-injection classifier and bad-actor checks, while response guardrails screen what the agent says for groundin…”
   • [claimed-docs] “every event lands in your analytics as a tracked outcome your QA team can review”
   • [claimed-docs] “Author adversarial scenarios such as false authority claims, mid-conversation goal switches and prompt injection attempts, then run them as …”
   • [claimed-docs] “Enforce required disclosures and data handling standards through built-in guardrails that balance conversational flexibility with structured…”

   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
   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

   Lorikeet—none0/10

   Lorikeet's documented model is autonomous resolution with post-hoc QA (Coach reviews 100% of tickets after the fact) and escalation to humans only when the AI can't resolve a case, not a pre-send draft-for-approval workflow. No evidence describes a mode where every agent reply is queued for human sign-off before reaching a customer. Missing for 10: any mention of a draft/approve workflow, human-in-the-loop send gating, or a configurable 'supervised' launch mode.

   • [claimed-docs] “When the AI concierge can't resolve a conversation, a human agent steps in directly inside the same platform”
   • [claimed-docs] “Incoming messages pass a prompt-injection classifier and bad-actor checks, while response guardrails screen what the agent says for groundin…”
   • [claimed-docs] “Coach’s Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “When a case requires a specialist or regulatory review, it escalates with full interaction history so your team picks up mid-conversation.”

   proof ↗⚑ 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 drawn
   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

   Lorikeet~partialclaimed5/10

   Lorikeet documents topic-scoped deployment ("deploy on the topics the agent is trained for and leave the rest with your team"), workflow-scoped tool access with hard execution caps enforced in code, and escalation paths for regulated/specialist cases with full history handoff — all consistent with restricting the agent from acting on sensitive topics. However, there's no explicit documentation of a UI/config for support-ops to designate specific topics (e.g., legal threats, cancellations, security) as strictly human-only with enforced non-freelancing. Missing for 10: explicit topic-tagging/human-only designation feature, confirmation that agent cannot even attempt those topics (vs. escalating mid-conversation), and independent verification of this behavior in production.

   • [claimed-docs] “deploy on the topics the agent is trained for and leave the rest with your team”
   • [claimed-docs] “Lorikeet coordinates a team of specialist agents, with pockets of determinism for regulated steps, to handle multi-party, multi-system workf…”
   • [claimed-docs] “Customer isolation, server-side identity validation, workflow-scoped tool access and hard execution caps are enforced in code rather than pr…”
   • [claimed-docs] “When a case requires a specialist or regulatory review, it escalates with full interaction history so your team picks up mid-conversation.”
   • [claimed-docs] “Enforce required disclosures and data handling standards through built-in guardrails that balance conversational flexibility with structured…”

   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 drawn
   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

   Lorikeet~partialclaimed5/10

   Lorikeet claims analytics tracking 'resolution quality, customer satisfaction, revenue impact, and operational efficiency' and that 'every event lands in your analytics as a tracked outcome,' which loosely maps to resolution rate, CSAT, and cost metrics, but there is no explicit mention of a handoff-rate metric or a dedicated exec-facing dashboard view combining these four numbers. missing for 10: explicit handoff-rate metric, dedicated dashboard UI/screenshot evidence, cost-per-resolution specifics, and independent corroboration of dashboard usability.

   • [claimed-docs] “Track resolution quality, customer satisfaction, revenue impact, and operational efficiency with industry-leading analytics.”
   • [claimed-docs] “Scale and optimize with conversational insights and analytics from Lorikeet Coach”
   • [claimed-docs] “every event lands in your analytics as a tracked outcome your QA team can review”

   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
   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

   Lorikeet—none0/10

   Evidence shows general analytics and quality-review features (e.g., 'Track resolution quality, customer satisfaction... with industry-leading analytics', Coach reviewing 100% of conversations) but no mention of topic clustering of conversations or proactive surfacing of emerging product issues before ticket-volume spikes. missing for 10: topic clustering of conversations, trend/anomaly detection for emerging issues, ticket-volume spike prediction or alerting.

   • [claimed-docs] “Track resolution quality, customer satisfaction, revenue impact, and operational efficiency with industry-leading analytics.”
   • [claimed-docs] “Scale and optimize with conversational insights and analytics from Lorikeet Coach”
   • [claimed-docs] “Coach's Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”

   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 Lorikeet
   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

   Lorikeet~partialclaimed5/10

   Docs claim Lorikeet works 'alongside your existing tools' and connects 'in seconds to your ticketing system' via integrations, implying helpdesk co-existence, but there is no specific evidence of named Zendesk/Salesforce/Intercom connectors or two-way ticket/context sync — the pack focuses on MCP client integrations (Claude, ChatGPT) rather than helpdesk platforms. missing for 10: named Zendesk/Salesforce/Intercom integration docs, evidence of bidirectional ticket sync, standalone-mode confirmation, independent corroboration of integration reliability.

   • [claimed-docs] “No migration required: Lorikeet works alongside your existing tools.”
   • [claimed-docs] “Connect Lorikeet in seconds to your ticketing system, knowledge base, and internal tools to seamlessly ingest data and take action for your …”
   • [claimed-docs] “When the AI concierge can't resolve a conversation, a human agent steps in directly inside the same platform”
   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”

   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 drawn
   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

   Lorikeet—none0/10

   Evidence shows Lorikeet ingests knowledge bases and can audit them for gaps/outdated articles, and connects to ticketing/knowledge sources 'in seconds', but there is no mention of scheduled or change-triggered re-syncing of sources without manual re-upload. missing for 10: explicit scheduled/webhook-based re-sync mechanism, evidence of automatic detection of source updates, any documentation of sync cadence or on-change triggers.

   • [claimed-docs] “Audit knowledge bases - analyze articles at scale, find gaps, and spot quality issues”
   • [claimed-docs] “Connect Lorikeet in seconds to your ticketing system, knowledge base, and internal tools to seamlessly ingest data and take action for your …”
   • [claimed-docs] “Audit your entire knowledge base for gaps, outdated articles, and quality issues”
   • [claimed-docs] “train the agent on your business context, brand guidelines, help docs and standard operating procedures”

   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 Lorikeet
   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

   Lorikeet✓fullclaimed8/10

   Lorikeet explicitly documents auditing knowledge bases at scale to find gaps and quality issues, and its simulation/replay tooling surfaces knowledge gaps by replaying historical tickets and synthetic scenarios to project resolution quality before deployment. Coach's Ticket Quality Score also reviews 100% of conversations to catch fumbled answers and feed fixes back. Missing for 10: independent/hands-on validation of the gap-detection accuracy and no explicit mention of detecting 'conflicting' content specifically (only gaps/outdated/quality issues).

   • [claimed-docs] “Audit knowledge bases - analyze articles at scale, find gaps, and spot quality issues”
   • [claimed-docs] “Audit your entire knowledge base for gaps, outdated articles, and quality issues”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps, then deploy on the topics …”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps”
   • [claimed-docs] “Coach's Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “Coach’s Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “Coach reviews 100% of tickets against your quality standards and turns findings into fixes, so your AI agent improves every week instead of …”

   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 drawn
   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

   Lorikeet—none0/10

   Lorikeet documents training its agent on 'business context, brand guidelines, help docs and standard operating procedures' (docs-25) and mentions 'grounding' as a guardrail check (docs-35), but there is no evidence that end-user-facing answers cite or display the specific article/source used to generate a response. Missing for 10: any documented citation/source-attribution UI or API in agent responses, and independent confirmation that answers reference specific knowledge-base articles.

   • [claimed-docs] “train the agent on your business context, brand guidelines, help docs and standard operating procedures”
   • [claimed-docs] “Incoming messages pass a prompt-injection classifier and bad-actor checks, while response guardrails screen what the agent says for groundin…”
   • [claimed-docs] “Audit your entire knowledge base for gaps, outdated articles, and quality issues”

   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 Lorikeet
   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

   Lorikeet✓fullclaimed7/10

   Lorikeet explicitly claims to 'seamlessly ingest data' from ticketing systems and knowledge bases (docs-13), to train the agent on 'business context, brand guidelines, help docs and standard operating procedures' (docs-25), to audit the entire knowledge base for gaps/outdated content (docs-3, docs-16), and to replay historical tickets to surface knowledge gaps (docs-21, docs-26) — covering help center, docs, and past tickets without manual re-authoring. Internal wikis are not explicitly named as a source type, and all evidence is first-party vendor documentation with no independent/hands-on corroboration of ingestion working end-to-end. Missing for 10: explicit mention of internal wiki ingestion, and independent verification of the ingestion pipeline's fidelity/accuracy.

   • [claimed-docs] “Connect Lorikeet in seconds to your ticketing system, knowledge base, and internal tools to seamlessly ingest data and take action for your …”
   • [claimed-docs] “train the agent on your business context, brand guidelines, help docs and standard operating procedures”
   • [claimed-docs] “Audit knowledge bases - analyze articles at scale, find gaps, and spot quality issues”
   • [claimed-docs] “Audit your entire knowledge base for gaps, outdated articles, and quality issues”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps, then deploy on the topics …”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps”

   proof ↗⚑ flag

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

   weight 2 · round to Lorikeet
   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

   Lorikeet~partialprobed6/10

   Lorikeet's MCP server exposes a broad set of actions also available in the UI—diagnosing tickets, auditing knowledge bases, building/testing workflows, running simulations—letting AI-native users act via Claude/ChatGPT/Codex instead of the dashboard. However, there's no evidence of a general-purpose REST/GraphQL API or explicit claim of full UI/API parity (e.g., outbound campaign management, guardrail configuration, analytics dashboards aren't shown as MCP-accessible). Missing for 10: explicit full API parity statement, broader API coverage beyond MCP-listed actions, independent confirmation of completeness.

   • [claimed-docs] “The Lorikeet MCP server lets you interact with your Lorikeet account directly from Claude Code, Claude.ai, ChatGPT, and Codex.”
   • [claimed-docs] “Diagnose tickets - trace workflow execution and identify root causes”
   • [claimed-docs] “Audit knowledge bases - analyze articles at scale, find gaps, and spot quality issues”
   • [claimed-docs] “Build workflows - create and iterate on workflows using natural language”
   • [claimed-docs] “Test tools - run and validate tool configurations directly from your AI assistant”
   • [claimed-docs] “Explore your setup - inspect workflows, tools, and integrations”
   • [claimed-docs] “Run simulations - test workflows against different customer scenarios”
   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”
   • [probe] “official MCP server documented at https://docs.lorikeetcx.ai/mcp/mcp-server”

   proof ↗⚑ flag

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

   weight 3 · round drawn
   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

   Lorikeet—none0/10

   No evidence pack items mention data export, open data formats, or account/data portability for leaving the platform; the docs cover MCP integrations, workflows, simulations, and guardrails but nothing about exporting or migrating data out. Missing for 10: any documentation of export functionality, supported open formats (CSV/JSON), or data portability/exit process.

   ⚑ 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 Lorikeet
   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

   Lorikeet~partialclaimed3/10

   The only evidence touching pricing-economics is the refund policy tying billing to Coach's quality score ('When Coach gives a conversation a bad score, we refund the AI portion of that interaction'), which shows some outcome-linkage in billing, but there is no published price list, per-resolution rate, cap structure, or self-serve pricing page. Missing for 10: a published price sheet or rate card, explicit per-resolution cost, spend/volume caps, and any evidence pricing is transparent versus custom enterprise quoting.

   • [claimed-docs] “When Coach gives a conversation a bad score, we refund the AI portion of that interaction.”

   proof ↗⚑ flag

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

   weight 2 · round drawn
   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

   Lorikeet—none0/10

   The evidence covers compliance/trust items (Vanta certifications, zero-data-retention with model vendors) but no mention of data residency or region-selection options for storage. Missing for 10: any documentation of regional data storage choices, residency guarantees, or data localization controls.

   ⚑ flag

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

   weight 3 · round to Lorikeet
   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

   Lorikeet✓fullclaimed7/10

   Lorikeet explicitly states zero-data-retention agreements with all model vendors and no fine-tuning on customer data, directly addressing the story's request to prevent data from being used for AI training. This is backed by independently verified trust/compliance reports on their Vanta Trust Center. Missing for 10: independent hands-on verification or third-party audit confirmation of this specific claim, and no detail on user-level opt-out controls or granularity of enforcement.

   • [claimed-docs] “Zero-data-retention agreements with all model vendors and no fine-tuning on customer data.”
   • [claimed-docs] “All four are independently verified, published on our public Vanta Trust Center with reports downloadable under NDA, and refreshed annually.”

   proof ↗⚑ flag

3. ai-native userControl data retention and deletion

   weight 2 · round to Lorikeet
   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

   Lorikeet~partialclaimed4/10

   Lorikeet mentions zero-data-retention agreements with model vendors and no fine-tuning on customer data, plus SOC2-style independently verified reports on a trust center, which touches data retention posture at the vendor-model level. However, there is no evidence of user-facing controls letting an AI-native user configure or request deletion/retention of their own conversation or account data within Lorikeet itself. Missing for 10: explicit customer-data deletion/export controls, retention period configuration, and user-initiated deletion workflows.

   • [claimed-docs] “All four are independently verified, published on our public Vanta Trust Center with reports downloadable under NDA, and refreshed annually.”
   • [claimed-docs] “Zero-data-retention agreements with all model vendors and no fine-tuning on customer data.”

   proof ↗⚑ flag

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

   weight 2 · round drawn
   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

   Lorikeet—none0/10

   Lorikeet is a customer-support AI platform, and telemetry opt-out for the product itself is a fair privacy-posture question, but none of the evidence mentions any telemetry/usage-tracking opt-out mechanism for users of the product; it only discusses data retention with model vendors and consent handling for outbound customer messaging, which is unrelated to product telemetry opt-out.

   • [claimed-docs] “All four are independently verified, published on our public Vanta Trust Center with reports downloadable under NDA, and refreshed annually.”
   • [claimed-docs] “Zero-data-retention agreements with all model vendors and no fine-tuning on customer data.”

   proof ↗⚑ 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 Lorikeet
   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

   Lorikeet✓fullclaimed7/10

   Lorikeet's docs describe connecting to ticketing systems, knowledge bases, and internal tools/APIs to 'ingest data and take action for your customers' (lorikeet-docs-13, lorikeet-docs-28, lorikeet-docs-29), with concrete examples like testing a 'get-order-status' tool with a real order ID (lorikeet-docs-19) and financial-services use cases like disputes/loan inquiries requiring account-specific data (lorikeet-docs-43, lorikeet-docs-44). This shows the agent is designed to pull and act on live customer data rather than just static help content. missing for 10: independent/hands-on verification that responses actually reflect real-time account state in production, and more detail on latency/freshness guarantees for live data lookups.

   • [claimed-docs] “Connect Lorikeet in seconds to your ticketing system, knowledge base, and internal tools to seamlessly ingest data and take action for your …”
   • [claimed-docs] “Test the 'get-order-status' tool with order ID 98765 and check if the response matches what we expect.”
   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”
   • [claimed-docs] “Build your customer Concierge to take actions inside the tools your team already uses”
   • [claimed-docs] “Automate disputes, loan inquiries, payments and compliance cases across every customer channel”
   • [claimed-docs] “When a case requires a specialist or regulatory review, it escalates with full interaction history so your team picks up mid-conversation.”

   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 Lorikeet
   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

   Lorikeet~partialclaimed6/10

   Lorikeet's docs describe workflows that coordinate multi-step, multi-system troubleshooting (e.g. 'coordinates a team of specialist agents... to handle multi-party, multi-system workflows end-to-end' and 'diagnose tickets by tracing workflow execution') rather than single canned replies, and its agent resolves issues 'end-to-end' across channels. However, there is no explicit documentation showing the agent proactively asking clarifying questions mid-conversation or examples of dialogue-based troubleshooting turns. Missing for 10: explicit examples/transcripts of clarifying-question behavior, documentation describing conversational back-and-forth troubleshooting logic rather than just workflow/tool orchestration.

   • [claimed-docs] “Lorikeet coordinates a team of specialist agents, with pockets of determinism for regulated steps, to handle multi-party, multi-system workf…”
   • [claimed-docs] “one agent that resolves issues end-to-end across chat, email, voice and SMS”
   • [claimed-docs] “Diagnose tickets by tracing workflow execution and identifying root causes”
   • [claimed-docs] “Build, edit, and deploy workflows using natural language”
   • [claimed-docs] “The agent takes action through your APIs and MCP servers inside natural-language and deterministic workflows”

   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 Lorikeet
   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

   Lorikeet~partialclaimed6/10

   Lorikeet's docs repeatedly claim end-to-end resolution (chat, email, voice, SMS), with escalation to humans when it can't resolve, plus QA scoring (Ticket Quality Score) and analytics tracking 'resolution quality' as an outcome metric, and even a refund-on-bad-score mechanism tied to quality. However, all evidence is vendor-authored marketing/docs; there are no independent benchmarks, customer case studies, or hard resolution-rate numbers (e.g., % of conversations fully resolved) to substantiate the claims. Missing for 10: independent/third-party resolution-rate data, customer-reported metrics, and clear definition/measurement methodology distinguishing true resolution from deflection.

   • [claimed-docs] “one agent that resolves issues end-to-end across chat, email, voice and SMS”
   • [claimed-docs] “resolve customer problems end-to-end across their lifecycle - via phone, SMS, chat, email, and WhatsApp”
   • [claimed-docs] “When the AI concierge can't resolve a conversation, a human agent steps in directly inside the same platform”
   • [claimed-docs] “Track resolution quality, customer satisfaction, revenue impact, and operational efficiency with industry-leading analytics.”
   • [claimed-docs] “Coach’s Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “When Coach gives a conversation a bad score, we refund the AI portion of that interaction.”
   • [claimed-docs] “Coach's Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”

   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
   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

   Lorikeet~partialclaimed6/10

   Lorikeet docs show the agent can be trained on 'business context, brand guidelines, help docs and standard operating procedures' (lorikeet-docs-25), and Coach's Ticket Quality Score reviews 100% of conversations against quality standards to catch drift (lorikeet-docs-9, lorikeet-docs-38), supporting brand-voice control and consistency monitoring. However, there is no explicit evidence of multi-language tone consistency or dedicated brand-voice/style configuration tooling beyond general training inputs. Missing for 10: explicit multilingual consistency support, dedicated tone/voice configuration UI, and independent evidence of voice consistency across topics/languages.

   • [claimed-docs] “train the agent on your business context, brand guidelines, help docs and standard operating procedures”
   • [claimed-docs] “Coach's Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “Coach’s Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “deploy on the topics the agent is trained for and leave the rest with your team”

   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 drawn
   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

   Lorikeet✓fullclaimed8/10

   Lorikeet's Coach product directly addresses this story: it reviews 100% of conversations against quality standards (not just samples), scores them via a Ticket Quality Score, flags failures (bad scores trigger refunds), and turns findings into fixes that feed back into the agent so it 'improves every week instead of drifting.' Simulations complement this with batch testing and adversarial scenario scoring tied to guardrails and analytics tracking. Missing for 10: independent/hands-on verification of the review loop in practice and more detail on how flagged failures are triaged/assigned to human reviewers.

   • [claimed-docs] “Coach's Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “Coach reviews 100% of tickets against your quality standards and turns findings into fixes, so your AI agent improves every week instead of …”
   • [claimed-docs] “Coach’s Ticket Quality Score reviews 100% of conversations against your quality standards, AI and human alike, replacing manual spot checks …”
   • [claimed-docs] “When Coach gives a conversation a bad score, we refund the AI portion of that interaction.”
   • [claimed-docs] “Coach can implement improvements on your behalf, or make suggestions for you to action yourself.”
   • [claimed-docs] “every event lands in your analytics as a tracked outcome your QA team can review”
   • [claimed-docs] “Generate simulations straight from your actual tickets and run them in bulk batches, so every workflow change is tested against the conversa…”
   • [claimed-docs] “Author adversarial scenarios such as false authority claims, mid-conversation goal switches and prompt injection attempts, then run them as …”

   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 Lorikeet
   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

   Lorikeet✓fullclaimed9/10

   Lorikeet's Simulations product directly supports this story: it generates simulations from actual historical tickets and runs them in bulk batches before workflow changes go live, with side-by-side batch comparisons and per-conversation drill-downs, plus authored adversarial/guardrail scenarios to pre-test against tricky real-world behavior. Docs also describe replaying historical tickets and synthetic scenarios in bulk to project resolution quality before deploying on trained topics. Missing for 10: independent/hands-on corroboration beyond first-party docs.

   • [claimed-docs] “Generate simulations straight from your actual tickets and run them in bulk batches, so every workflow change is tested against the conversa…”
   • [claimed-docs] “Side-by-side batch comparisons show how a workflow edit changed outcomes across hundreds of scenarios, with per-conversation drill-downs whe…”
   • [claimed-docs] “Author adversarial scenarios such as false authority claims, mid-conversation goal switches and prompt injection attempts, then run them as …”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps, then deploy on the topics …”
   • [claimed-docs] “Replay historical tickets and synthetic scenarios in bulk to see projected resolution quality and knowledge gaps”
   • [claimed-docs] “Run simulations - test workflows against different customer scenarios”

   proof ↗⚑ flag