HonestClaw · What is Agent Readiness? · How we score
Developing. Agents can find parts of this site, but key capabilities are hard to reach.
HonestClaw measured how well confident-ai.com's site and APIs work for AI agents — across 9 dimensions and 45 automated checks. Score: 51/100 (Developing).
Scanned · scanner v1.67.0 · 44 of 45 checks conclusive.
All 8 scored areas have measured gaps. Access & rendering is furthest along; Reliability & recovery has the largest measured point gap.
81 additional observations provide context and do not affect the score.
Reachable entry points
No DNS-AID index found at _index._agents.confident-ai.com.
These are recorded points before the total is normalized to 100.
Public pages; interfaces when present
Clear entry points help agents find the intended site or interface. Multiple hostnames or directory formats may describe the same destination.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
API Catalog found at https://confident-ai.com/.well-known/api-catalog (application/linkset+json).
These are recorded points before the total is normalized to 100.
Public pages; interfaces when present
Clear entry points help agents find the intended site or interface. Multiple hostnames or directory formats may describe the same destination.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No RFC 8288 Link headers found on any probed response. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$; Disallow: /api/*).
These are recorded points before the total is normalized to 100.
Public pages; interfaces when present
Clear entry points help agents find the intended site or interface. Multiple hostnames or directory formats may describe the same destination.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
No non-owned /.well-known/ descriptors found.
These are recorded points before the total is normalized to 100.
Public pages; interfaces when present
Clear entry points help agents find the intended site or interface. Multiple hostnames or directory formats may describe the same destination.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
An agent-relevant subdomain is live and serves its expected content kind (docs.confident-ai.com).
These are recorded points before the total is normalized to 100.
Public pages; interfaces when present
Clear entry points help agents find the intended site or interface. Multiple hostnames or directory formats may describe the same destination.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
Root and www converge on www.confident-ai.com, with a matching canonical tag.
These are recorded points before the total is normalized to 100.
Public pages; interfaces when present
Clear entry points help agents find the intended site or interface. Multiple hostnames or directory formats may describe the same destination.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
https://www.confident-ai.com/sitemap.xml is a dated sitemap whose URLs include docs/pricing/api surfaces.
These are recorded points before the total is normalized to 100.
Public pages; interfaces when present
Clear entry points help agents find the intended site or interface. Multiple hostnames or directory formats may describe the same destination.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Attributable ecosystem discovery
Off-domain evidence
official MCP Registry search page could not be read or returned an invalid response; absence was not inferred.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Only the relevant verified distribution surface
An attributable directory entry can help agents discover a supported interface. Popularity, company size and directory count do not establish that the interface works.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Off-domain evidence
The fetched GitHub agent-rules repository search page returned no listing attributable to the scanned company.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Only the relevant verified distribution surface
An attributable directory entry can help agents discover a supported interface. Popularity, company size and directory count do not establish that the interface works.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Off-domain evidence
The fetched GitHub agent-plugin repository search page returned no listing attributable to the scanned company.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Only the relevant verified distribution surface
An attributable directory entry can help agents discover a supported interface. Popularity, company size and directory count do not establish that the interface works.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Off-domain evidence
skills.sh returned 6 listing(s) attributable to the company. Presence alone is capped at detected.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Only the relevant verified distribution surface
An attributable directory entry can help agents discover a supported interface. Popularity, company size and directory count do not establish that the interface works.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Off-domain evidence
6 attributable skills.sh listing(s) expose a named source and install metadata. This records directory metadata, not a claim of editorial quality, and is capped at detected.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Only the relevant verified distribution surface
An attributable directory entry can help agents discover a supported interface. Popularity, company size and directory count do not establish that the interface works.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Off-domain evidence
official MCP Registry search page could not be read or returned an invalid response; absence was not inferred.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Only the relevant verified distribution surface
An attributable directory entry can help agents discover a supported interface. Popularity, company size and directory count do not establish that the interface works.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Off-domain evidence
ChatGPT Plugins Directory page could not be read or returned an invalid response; absence was not inferred.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Only the relevant verified distribution surface
An attributable directory entry can help agents discover a supported interface. Popularity, company size and directory count do not establish that the interface works.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Commerce distribution
Off-domain evidence
The fetched x402 Bazaar discovery search returned no listing attributable to the scanned company.
Useful context, outside the readiness score. Adoption, payment support and retrieval cost are not readiness quality by themselves.
Payment-enabled products
A payment directory is useful context. Making it a general ARS requirement would indirectly score a business-model choice.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Equivalent readable representations
A markdown twin is published at https://www.confident-ai.com/docs.md alongside the docs HTML surface (https://www.confident-ai.com/docs).
These are recorded points before the total is normalized to 100.
Only representations the site offers; all sites can satisfy readable-content outcome with HTML
Headers, routes and frontmatter support one readable representation. Plain HTML must be able to receive full content-access credit.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No reachable candidate HTML page declared a text/markdown alternate.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Only representations the site offers; all sites can satisfy readable-content outcome with HTML
Headers, routes and frontmatter support one readable representation. Plain HTML must be able to receive full content-access credit.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
A direct Markdown representation is available at https://www.confident-ai.com/docs.md.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Only representations the site offers; all sites can satisfy readable-content outcome with HTML
Headers, routes and frontmatter support one readable representation. Plain HTML must be able to receive full content-access credit.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No reachable candidate page served Markdown to the scanner agent without an Accept header.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Only representations the site offers; all sites can satisfy readable-content outcome with HTML
Headers, routes and frontmatter support one readable representation. Plain HTML must be able to receive full content-access credit.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
The page negotiates Markdown, but the response does not declare Vary: Accept; shared caches can mix HTML and Markdown variants.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Only representations the site offers; all sites can satisfy readable-content outcome with HTML
Headers, routes and frontmatter support one readable representation. Plain HTML must be able to receive full content-access credit.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No readable Markdown representation began with YAML frontmatter.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Only representations the site offers; all sites can satisfy readable-content outcome with HTML
Headers, routes and frontmatter support one readable representation. Plain HTML must be able to receive full content-access credit.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
Content available to the tested client
Readable content is extractable without JavaScript on 2 probed surface(s) with a strong main-content ratio (90% on the homepage).
These are recorded points before the total is normalized to 100.
Public pages; distinguish HTTP and browser clients
Critical content available in the response is a direct prerequisite for the tested client. An extraction ratio alone does not prove that the needed facts were captured.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Retrieval cost
A manageable sitemap (968 page URLs) helps agents estimate crawl cost.
Useful context, outside the readiness score. Adoption, payment support and retrieval cost are not readiness quality by themselves.
Measured retrieval path
Tokens, bytes and elapsed time explain retrieval cost and tradeoffs. They are context and do not affect the readiness score.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Accessibility-tree operability signals
Accessibility-tree probe unavailable (deep_scan_only): Accessibility-tree checks are deep-scan only to avoid always-on browser probe cost.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Deep scans with an available browser accessibility tree snapshot
Accessible names, form labels, document structure and native semantics are direct operability signals for agents acting through accessibility trees. The current reading is deep-scan gated and intentionally unweighted pending calibration.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Accessibility-tree probe unavailable (deep_scan_only): Accessibility-tree checks are deep-scan only to avoid always-on browser probe cost.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Deep scans with an available browser accessibility tree snapshot
Accessible names, form labels, document structure and native semantics are direct operability signals for agents acting through accessibility trees. The current reading is deep-scan gated and intentionally unweighted pending calibration.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Accessibility-tree probe unavailable (deep_scan_only): Accessibility-tree checks are deep-scan only to avoid always-on browser probe cost.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Deep scans with an available browser accessibility tree snapshot
Accessible names, form labels, document structure and native semantics are direct operability signals for agents acting through accessibility trees. The current reading is deep-scan gated and intentionally unweighted pending calibration.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Accessibility-tree probe unavailable (deep_scan_only): Accessibility-tree checks are deep-scan only to avoid always-on browser probe cost.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Deep scans with an available browser accessibility tree snapshot
Accessible names, form labels, document structure and native semantics are direct operability signals for agents acting through accessibility trees. The current reading is deep-scan gated and intentionally unweighted pending calibration.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Reachable critical content
Static same-site links led to substantive HTML content two crawl hops from the homepage.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Sites with linked public content
Bounded path traversal can explain missing content. Absolute clicks from the homepage is not a validated universal quality scale.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Observed client access
4 of 5 bot-detection probes passed. ChatGPT-User=degraded, ClaudeBot=clear, Google-Extended=clear, PerplexityBot=clear, DeepSeekBot=clear
A limited scanner observation, not proof of access from real agent infrastructure. It remains unscored until the probe can support that claim.
Named client and network context; publisher-permitted retrieval
Current probes identify as HonestClaw with agent-token markers. They cannot verify real vendor access. A bot-protection vendor name alone is not a failure.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
4 of 5 agent crawler probes passed. ChatGPT-User=0, ClaudeBot=200, Google-Extended=200, PerplexityBot=200, DeepSeekBot=200
A limited scanner observation, not proof of access from real agent infrastructure. It remains unscored until the probe can support that claim.
Named client and network context; publisher-permitted retrieval
Current probes identify as HonestClaw with agent-token markers. They cannot verify real vendor access. A bot-protection vendor name alone is not a failure.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Clear and consistent publisher policy
Could not conclusively check for robots.txt (unreachable, rate-limited, or behind auth).
Unverified points are excluded from the score denominator and reduce observed coverage.
Published policy for the relevant use: indexing, training or user retrieval
Different policy syntaxes and business permissions are not independent readiness capabilities. Blocking training should not imply that an agent cannot use a public API.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
robots.txt is readable but declares no RSL License directive.
These are recorded points before the total is normalized to 100.
Published policy for the relevant use: indexing, training or user retrieval
Different policy syntaxes and business permissions are not independent readiness capabilities. Blocking training should not imply that an agent cannot use a public API.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No TDMRep policy found at /.well-known/tdmrep.json.
These are recorded points before the total is normalized to 100.
Published policy for the relevant use: indexing, training or user retrieval
Different policy syntaxes and business permissions are not independent readiness capabilities. Blocking training should not imply that an agent cannot use a public API.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No ai.txt file found (HTML pages at .txt paths do not qualify).
These are recorded points before the total is normalized to 100.
Published policy for the relevant use: indexing, training or user retrieval
Different policy syntaxes and business permissions are not independent readiness capabilities. Blocking training should not imply that an agent cannot use a public API.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
robots.txt declares explicit rules for 14 AI bot(s): gptbot, oai-searchbot, chatgpt-user, perplexitybot, perplexity-user, claudebot, claude-web, anthropic-ai, google-extended, applebot-extended, ccbot, cohere-ai, amazonbot, meta-externalagent.
These are recorded points before the total is normalized to 100.
Published policy for the relevant use: indexing, training or user retrieval
Different policy syntaxes and business permissions are not independent readiness capabilities. Blocking training should not imply that an agent cannot use a public API.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
robots.txt declares 3 recognized Content Signal(s) with a valid value: search, ai-input, ai-train.
These are recorded points before the total is normalized to 100.
Published policy for the relevant use: indexing, training or user retrieval
Different policy syntaxes and business permissions are not independent readiness capabilities. Blocking training should not imply that an agent cannot use a public API.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Usable API reference and examples
An API reference page exists at https://www.confident-ai.com/docs/reference, but it doesn't show parameters with example requests and responses.
These are recorded points before the total is normalized to 100.
A documented API or tool interface
Reference depth and valid fenced examples contribute to the same documentation outcome. Parsing an example is weaker than executing it successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No fenced code blocks were found across 6 readable Markdown document(s).
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
A documented API or tool interface
Reference depth and valid fenced examples contribute to the same documentation outcome. Parsing an example is weaker than executing it successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
Usable documentation entry and references
No llms-full.txt file found (HTML pages at .txt paths do not qualify).
These are recorded points before the total is normalized to 100.
Public documentation; llms files optional
An index, full text and modular files offer different routes to documentation. Useful content and working references matter more than the number of formats.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
A curated llms.txt index is published at https://www.confident-ai.com/llms.txt: 122 links across 9 sections.
These are recorded points before the total is normalized to 100.
Public documentation; llms files optional
An index, full text and modular files offer different routes to documentation. Useful content and working references matter more than the number of formats.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The scoped llms.txt lists 800 resource(s) inside /docs.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Public documentation; llms files optional
An index, full text and modular files offer different routes to documentation. Useful content and working references matter more than the number of formats.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
2/5 sampled on-domain llms.txt links resolved to real, non-shell content.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Public documentation; llms files optional
An index, full text and modular files offer different routes to documentation. Useful content and working references matter more than the number of formats.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Machine-readable facts
No machine-readable pricing endpoint found (HTML pricing pages do not qualify). Blocked by robots.txt for a generic compliant agent (Disallow: /api/*).
These are recorded points before the total is normalized to 100.
Facts relevant to the product; pricing only when there is a price to disclose
Structured facts can help agents interpret a page. Markup volume and the number of schema types alone do not establish accuracy, consistency or task success.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Observed 8 JSON-LD script(s); 8 parsed as JSON-LD objects.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Facts relevant to the product; pricing only when there is a price to disclose
Structured facts can help agents interpret a page. Markup volume and the number of schema types alone do not establish accuracy, consistency or task success.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Observed 11 resolvable identifier(s) and 4 verified entity link(s).
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Facts relevant to the product; pricing only when there is a price to disclose
Structured facts can help agents interpret a page. Markup volume and the number of schema types alone do not establish accuracy, consistency or task success.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Observed 10 Schema.org type(s): answer, breadcrumblist, faqpage, imageobject, listitem, organization, person, postaladdress, question, website; 4 meaningfully populated.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Facts relevant to the product; pricing only when there is a price to disclose
Structured facts can help agents interpret a page. Markup volume and the number of schema types alone do not establish accuracy, consistency or task success.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Document freshness evidence
Freshness signals present: ETag header, sitemap lastmod.
These are recorded points before the total is normalized to 100.
Resources whose currency can be assessed
A timestamp or ETag supports freshness provenance; it does not by itself prove factual accuracy or current behavior.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Usable agent guidance
No v0.2.0 index could be read.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Only a declared agent-guide surface
Alternative guide files may help agents locate instructions; file presence is supporting evidence and must not be required in every format.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No usable Agent Skills discovery index was found.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Only a declared agent-guide surface
Alternative guide files may help agents locate instructions; file presence is supporting evidence and must not be required in every format.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
The guidance is structured, but lacks both specific when-to-use and invocation guidance.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Only a declared agent-guide surface
Alternative guide files may help agents locate instructions; file presence is supporting evidence and must not be required in every format.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Complete and usable operation contracts
No data export or bulk API documentation found at the standard docs locations an agent would check.
These are recorded points before the total is normalized to 100.
Detected interface; pagination only for collections, async only for long-running jobs, batch only for multi-item operations
Usable schemas and operation patterns can reduce ambiguity for clients. More endpoints or constraints alone do not establish that an operation is easier to use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The spec at https://www.confident-ai.com/docs/openapi.yaml declares typed, consistent response schemas (99% of operations) with a pagination pattern and a structured error envelope. (source: spec)
These are recorded points before the total is normalized to 100.
Detected interface; pagination only for collections, async only for long-running jobs, batch only for multi-item operations
Usable schemas and operation patterns can reduce ambiguity for clients. More endpoints or constraints alone do not establish that an operation is easier to use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The spec declares a typed pagination parameter or response continuation field, but not a complete cursor round trip.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected interface; pagination only for collections, async only for long-running jobs, batch only for multi-item operations
Usable schemas and operation patterns can reduce ambiguity for clients. More endpoints or constraints alone do not establish that an operation is easier to use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The POST /v1/organization/audit-logs/exports operation declares 202 Accepted with job tracking fields (id, status).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected interface; pagination only for collections, async only for long-running jobs, batch only for multi-item operations
Usable schemas and operation patterns can reduce ambiguity for clients. More endpoints or constraints alone do not establish that an operation is easier to use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
A batch/bulk operation is named, but no array request body is declared.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected interface; pagination only for collections, async only for long-running jobs, batch only for multi-item operations
Usable schemas and operation patterns can reduce ambiguity for clients. More endpoints or constraints alone do not establish that an operation is easier to use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
All 198 operations have unique function-safe operationIds; 198/198 have typed inputs and 198/198 have descriptions or summaries.
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected interface; pagination only for collections, async only for long-running jobs, batch only for multi-item operations
Usable schemas and operation patterns can reduce ambiguity for clients. More endpoints or constraints alone do not establish that an operation is easier to use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
1581 parameter/request-body leaves across 198 operation(s): 374 machine-checkable (24%), 231 declare an enum (15%), 1521 carry a description (96%), 54 have neither a constraint nor a description (3%).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected interface; pagination only for collections, async only for long-running jobs, batch only for multi-item operations
Usable schemas and operation patterns can reduce ambiguity for clients. More endpoints or constraints alone do not establish that an operation is easier to use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Usable supported interface
A real API surface is declared at https://www.confident-ai.com/docs/openapi.yaml: the spec lists servers and paths. (source: spec)
These are recorded points before the total is normalized to 100.
Interface-enabled product; no requirement to implement every protocol
REST, GraphQL, MCP, SDKs and CLIs offer alternative or complementary access paths. The useful choice depends on the task and client; no single format serves every workflow.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No GraphQL transport surface found at conventional paths (GET-only; no introspection issued). Blocked by robots.txt for a generic compliant agent (Disallow: /api/*).
These are recorded points before the total is normalized to 100.
Interface-enabled product; no requirement to implement every protocol
REST, GraphQL, MCP, SDKs and CLIs offer alternative or complementary access paths. The useful choice depends on the task and client; no single format serves every workflow.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Valid OpenAPI 3.1.0 spec at https://www.confident-ai.com/docs/openapi.yaml: 198 operations, 374 schemas, 100% response coverage, servers declared.
These are recorded points before the total is normalized to 100.
Interface-enabled product; no requirement to implement every protocol
REST, GraphQL, MCP, SDKs and CLIs offer alternative or complementary access paths. The useful choice depends on the task and client; no single format serves every workflow.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
First-party documentation at https://www.confident-ai.com/docs/coding-agents/mcp identifies the MCP endpoint https://mcp.confident-ai.com/mcp and explains both its capabilities and how clients connect or authenticate.
These are recorded points before the total is normalized to 100.
Interface-enabled product; no requirement to implement every protocol
REST, GraphQL, MCP, SDKs and CLIs offer alternative or complementary access paths. The useful choice depends on the task and client; no single format serves every workflow.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
SDK libraries page at https://www.confident-ai.com/docs/integrations/third-party/claude-agent-sdk: 2 GitHub repos, 0 registry links, 1 install command (3 pointers).
These are recorded points before the total is normalized to 100.
Interface-enabled product; no requirement to implement every protocol
REST, GraphQL, MCP, SDKs and CLIs offer alternative or complementary access paths. The useful choice depends on the task and client; no single format serves every workflow.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No command-line tool was mentioned on any page checked.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Interface-enabled product; no requirement to implement every protocol
REST, GraphQL, MCP, SDKs and CLIs offer alternative or complementary access paths. The useful choice depends on the task and client; no single format serves every workflow.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Emerging interface declarations
No A2A Agent Card found at the standard well-known locations.
These are recorded points before the total is normalized to 100.
Only a positively identified matching protocol
Versioned interface declarations are observable signals. Broad adoption and cross-client utility are not established by declaration presence.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
An MCP Server Card at https://confident-ai.com/.well-known/mcp/server-card.json declares 2 remote endpoint(s) agents can connect to.
These are recorded points before the total is normalized to 100.
Only a positively identified matching protocol
Versioned interface declarations are observable signals. Broad adoption and cross-client utility are not established by declaration presence.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
WebMCP browser probing is not yet available in this scanner release.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Only a positively identified matching protocol
Versioned interface declarations are observable signals. Broad adoption and cross-client utility are not established by declaration presence.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Additional GraphQL contract evidence
No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response. GraphQL depth probe: No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$; Disallow: /api/*). This protocol's availability cannot be inferred from crawler policy.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
GraphQL and relevant operation shape; introspection optional
Additional contract evidence may substantiate an existing capability. Disabled production introspection must not create a penalty.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response. GraphQL depth probe: No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
GraphQL and relevant operation shape; introspection optional
Additional contract evidence may substantiate an existing capability. Disabled production introspection must not create a penalty.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response. GraphQL depth probe: No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
GraphQL and relevant operation shape; introspection optional
Additional contract evidence may substantiate an existing capability. Disabled production introspection must not create a penalty.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response. GraphQL depth probe: No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
GraphQL and relevant operation shape; introspection optional
Additional contract evidence may substantiate an existing capability. Disabled production introspection must not create a penalty.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
MCP discovery evidence
No MCP descriptor found at the standard locations an agent would check. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$).
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
MCP surface
An MCP descriptor helps clients locate a service. This additional observation does not earn separate points for evidence already represented by the scored MCP check.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Usable MCP tools and resources
Could not read live MCP tool listing from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool/resource capability
Live listings and contracts help agents discover tools and understand their inputs. These related observations remain unscored pending calibration of their contribution to successful tool use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP tool naming from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool/resource capability
Live listings and contracts help agents discover tools and understand their inputs. These related observations remain unscored pending calibration of their contribution to successful tool use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP tool descriptions from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool/resource capability
Live listings and contracts help agents discover tools and understand their inputs. These related observations remain unscored pending calibration of their contribution to successful tool use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP tool annotations from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool/resource capability
Live listings and contracts help agents discover tools and understand their inputs. These related observations remain unscored pending calibration of their contribution to successful tool use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP parameter schemas from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool/resource capability
Live listings and contracts help agents discover tools and understand their inputs. These related observations remain unscored pending calibration of their contribution to successful tool use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP parameter constraint form from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool capability
Machine-checkable constraints on tool inputs let something other than the model enforce them. The reading claims constraint form only, never that a server rejects bad input, and remains unscored pending calibration against the live MCP panel.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP resource listing from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool/resource capability
Live listings and contracts help agents discover tools and understand their inputs. These related observations remain unscored pending calibration of their contribution to successful tool use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP resource quality from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool/resource capability
Live listings and contracts help agents discover tools and understand their inputs. These related observations remain unscored pending calibration of their contribution to successful tool use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
3 MCP surfaces were declared; 0 completed live sessions. Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP server and advertised tool/resource capability
Live listings and contracts help agents discover tools and understand their inputs. These related observations remain unscored pending calibration of their contribution to successful tool use.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Compatible MCP transport
Could not read live MCP transport modernity from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Detected MCP transport and negotiated version
A compatible negotiated protocol version helps clients and servers communicate. An older supported transport can still be usable.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
MCP application UI
No adoption signal was detected, but endpoint discovery was incomplete or at least one declared MCP endpoint lacked readable list metadata, so site-level absence is unmeasurable. This metadata-only check did not call tools or read app resources.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected application UI surface
An embedded interface can support interactive workflows. Many tools work without a user interface, so its absence is not a general readiness failure.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No adoption signal was detected, but endpoint discovery was incomplete or at least one declared MCP endpoint lacked readable list metadata, so site-level absence is unmeasurable. This metadata-only check did not call tools or read app resources.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected application UI surface
An embedded interface can support interactive workflows. Many tools work without a user interface, so its absence is not a general readiness failure.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No adoption signal was detected, but endpoint discovery was incomplete or at least one declared MCP endpoint lacked readable list metadata, so site-level absence is unmeasurable. This metadata-only check did not call tools or read app resources.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected application UI surface
An embedded interface can support interactive workflows. Many tools work without a user interface, so its absence is not a general readiness failure.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Payment capabilities
Could not conclusively inspect published API specifications for Agentic Commerce Protocol checkout support. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$; Disallow: /api/*). This protocol's availability cannot be inferred from crawler policy.
Useful context, outside the readiness score. Adoption, payment support and retrieval cost are not readiness quality by themselves.
Payment-enabled business model
Payment support describes what the business offers. It must not make a research publication or free public service appear less ready.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not conclusively inspect published API specifications for ACP delegated-payment support. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$; Disallow: /api/*). This protocol's availability cannot be inferred from crawler policy.
Useful context, outside the readiness score. Adoption, payment support and retrieval cost are not readiness quality by themselves.
Payment-enabled business model
Payment support describes what the business offers. It must not make a research publication or free public service appear less ready.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not conclusively inspect the UCP profile and published API specifications for AP2. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$; Disallow: /api/*). This protocol's availability cannot be inferred from crawler policy.
Useful context, outside the readiness score. Adoption, payment support and retrieval cost are not readiness quality by themselves.
Payment-enabled business model
Payment support describes what the business offers. It must not make a research publication or free public service appear less ready.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The UCP discovery location could not be checked conclusively.
Useful context, outside the readiness score. Adoption, payment support and retrieval cost are not readiness quality by themselves.
Payment-enabled business model
Payment support describes what the business offers. It must not make a research publication or free public service appear less ready.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No machine-payment signal observed (no HTTP 402, no payment-required response header, and no /.well-known/x402.json manifest). Payment support cannot be ruled out from an unauthenticated probe. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$; Disallow: /api/*).
Useful context, outside the readiness score. Adoption, payment support and retrieval cost are not readiness quality by themselves.
Payment-enabled business model
Payment support describes what the business offers. It must not make a research publication or free public service appear less ready.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
NLWeb support
An NLWeb /ask operation was not observed in the bounded shared OpenAPI discovery surface.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected NLWeb service
Useful protocol observations; optional implementation is not universal readiness.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
NLWeb streaming support was not observed in the bounded shared OpenAPI discovery surface.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected NLWeb service
Useful protocol observations; optional implementation is not universal readiness.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
No valid HTTPS SchemaMap directive was found in robots.txt.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected NLWeb service
Useful protocol observations; optional implementation is not universal readiness.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
Safe integration environment
No sandbox or test environment documentation found at the standard docs locations an agent would check.
These are recorded points before the total is normalized to 100.
Operations whose side effects or onboarding require a safe test environment
A sandbox helps validate consequential integrations; it is not necessary for every public content site or read-only API.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Usable least-privilege authentication
Programmatic authentication is mentioned at https://www.confident-ai.com/docs/reference/api/v1/organization/api-keys/create-organization-api-key, but not on a page dedicated to documenting it.
These are recorded points before the total is normalized to 100.
Protected operations; anonymous access is valid when intended
Discovery, documentation and credential lifecycle jointly support a usable authorization flow. A public read-only service need not require OAuth.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
An authentication documentation page exists at https://www.confident-ai.com/docs/reference/api/authentication, but it doesn't name a specific method or show how a request is authenticated.
These are recorded points before the total is normalized to 100.
Protected operations; anonymous access is valid when intended
Discovery, documentation and credential lifecycle jointly support a usable authorization flow. A public read-only service need not require OAuth.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Credential management is documented at https://www.confident-ai.com/docs/reference/api/v1/organization/api-keys/create-organization-api-key, covering how to rotate or revoke credentials. To reach the top level, it would need rotation, revocation, and scoping/expiry all documented.
These are recorded points before the total is normalized to 100.
Protected operations; anonymous access is valid when intended
Discovery, documentation and credential lifecycle jointly support a usable authorization flow. A public read-only service need not require OAuth.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Agent-native auth discovery: a valid RFC 9728 PRM (via challenge) or a PRM plus a valid AS/OIDC descriptor.
These are recorded points before the total is normalized to 100.
Protected operations; anonymous access is valid when intended
Discovery, documentation and credential lifecycle jointly support a usable authorization flow. A public read-only service need not require OAuth.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
MCP authorization evidence
The endpoint advertises Bearer authentication with RFC 9728 resource metadata. Summary: weakest conclusive server; 2/3 declared endpoint(s) produced evidence, and every endpoint is reported separately. Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Protected HTTP MCP; OAuth/PKCE only for applicable flow
Authentication metadata and challenges help clients understand how to obtain access. Observing a declaration does not prove that credentials can be issued or used successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The challenge leads through protected-resource metadata to readable authorization-server metadata. Summary: weakest conclusive server; 2/3 declared endpoint(s) produced evidence, and every endpoint is reported separately. Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Protected HTTP MCP; OAuth/PKCE only for applicable flow
Authentication metadata and challenges help clients understand how to obtain access. Observing a declaration does not prove that credentials can be issued or used successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Authorization metadata advertises PKCE S256. Summary: weakest conclusive server; 2/3 declared endpoint(s) produced evidence, and every endpoint is reported separately. Live MCP probe: 10748 ms, 3 protocol request(s).
Measured, but not yet weighted. It adds no points and its absence subtracts none. Scoring requires calibration of reliability, applicability and overlap with existing checks.
Protected HTTP MCP; OAuth/PKCE only for applicable flow
Authentication metadata and challenges help clients understand how to obtain access. Observing a declaration does not prove that credentials can be issued or used successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Emerging authentication guidance
The root auth.md location could not be checked conclusively.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected emerging authentication scheme
An auth.md file is one possible guide. Reachable metadata and bounded simulation are not proof that credentials can be issued successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The root auth.md location could not be checked conclusively.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected emerging authentication scheme
An auth.md file is one possible guide. Reachable metadata and bounded simulation are not proof that credentials can be issued successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The root auth.md document does not satisfy the walkthrough criteria.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected emerging authentication scheme
An auth.md file is one possible guide. Reachable metadata and bounded simulation are not proof that credentials can be issued successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The Agent Auth discovery location could not be checked conclusively.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected emerging authentication scheme
An auth.md file is one possible guide. Reachable metadata and bounded simulation are not proof that credentials can be issued successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The Agent Auth discovery location could not be checked conclusively.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected emerging authentication scheme
An auth.md file is one possible guide. Reachable metadata and bounded simulation are not proof that credentials can be issued successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
API challenge headers could not be observed conclusively. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$; Disallow: /api/*). This protocol's availability cannot be inferred from crawler policy.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected emerging authentication scheme
An auth.md file is one possible guide. Reachable metadata and bounded simulation are not proof that credentials can be issued successfully.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
Observable outcomes and recovery
No webhook or event documentation found at the standard docs locations an agent would check.
These are recorded points before the total is normalized to 100.
Detected interface; idempotency for repeatable writes, webhooks only for event-driven workflows
Observable state and actionable errors help recovery. A status page or error schema does not prove successful repair.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No error documentation found at the standard docs locations an agent would check. Blocked by robots.txt for a generic compliant agent (Disallow: /api/*).
These are recorded points before the total is normalized to 100.
Detected interface; idempotency for repeatable writes, webhooks only for event-driven workflows
Observable state and actionable errors help recovery. A status page or error schema does not prove successful repair.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No retry or idempotency documentation found at the standard docs locations an agent would check.
These are recorded points before the total is normalized to 100.
Detected interface; idempotency for repeatable writes, webhooks only for event-driven workflows
Observable state and actionable errors help recovery. A status page or error schema does not prove successful repair.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No health or status endpoint found. Blocked by robots.txt for a generic compliant agent (Disallow: /api/*).
These are recorded points before the total is normalized to 100.
Detected interface; idempotency for repeatable writes, webhooks only for event-driven workflows
Observable state and actionable errors help recovery. A status page or error schema does not prove successful repair.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
The missing-page probe returned HTTP 0; missing-page behavior is inconclusive.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Detected interface; idempotency for repeatable writes, webhooks only for event-driven workflows
Observable state and actionable errors help recovery. A status page or error schema does not prove successful repair.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP error handling from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Detected interface; idempotency for repeatable writes, webhooks only for event-driven workflows
Observable state and actionable errors help recovery. A status page or error schema does not prove successful repair.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Actionable limits
A rate-limit documentation page exists at https://www.confident-ai.com/docs/settings/project/ai-connections/throttling-retries, but it doesn't state a concrete limit alongside a rate-limit response header.
These are recorded points before the total is normalized to 100.
An interface with quotas or throttling
Published limits and response headers help clients pace requests and recover from throttling. They describe related aspects of the same service constraint.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No confirmed live API response was available for rate-limit header inspection. Blocked by robots.txt for a generic compliant agent (Disallow: *.json$; Disallow: /api/*).
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
An interface with quotas or throttling
Published limits and response headers help clients pace requests and recover from throttling. They describe related aspects of the same service constraint.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
Change and compatibility information
The API version lifecycle is documented at https://www.confident-ai.com/docs/reference/api/v1/datasets/dataset-versions/get-dataset-versions, including deprecation/sunset. (source: docs)
These are recorded points before the total is normalized to 100.
Maintained interface or evolving product
Version identifiers and change records help clients manage compatibility as an interface evolves.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
A changelog is published at https://www.confident-ai.com/docs/changelog/self-hosting with multiple recent dated entries, indicating it is actively maintained.
These are recorded points before the total is normalized to 100.
Maintained interface or evolving product
Version identifiers and change records help clients manage compatibility as an interface evolves.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Additional GraphQL recovery evidence
No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response. GraphQL depth probe: No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
GraphQL service; schema evidence obtainable under permitted scope
GraphQL error definitions and evolution policies help clients recover and manage compatibility. Disabled production introspection does not establish a defect.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response. GraphQL depth probe: No GraphQL endpoint was observed in a scanner-fetched declaration or GraphQL-shaped response.
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
GraphQL service; schema evidence obtainable under permitted scope
GraphQL error definitions and evolution policies help clients recover and manage compatibility. Disabled production introspection does not establish a defect.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
Attributable identity and consistent claims
No security.txt found at /.well-known/security.txt or /security.txt.
These are recorded points before the total is normalized to 100.
Public company/service identity; MCP identity only for MCP
These establish provenance, disclosure routes and consistency. They cannot certify prompt-injection resistance or safe transactions.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
At least one legal policy page is discoverable.
These are recorded points before the total is normalized to 100.
Public company/service identity; MCP identity only for MCP
These establish provenance, disclosure routes and consistency. They cannot certify prompt-injection resistance or safe transactions.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No public evidence URL recorded for this check
Homepage exposes three org-identity signal types (legal_name, address, social).
These are recorded points before the total is normalized to 100.
Public company/service identity; MCP identity only for MCP
These establish provenance, disclosure routes and consistency. They cannot certify prompt-injection resistance or safe transactions.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Homepage metadata (canonical, Open Graph, schema) is present and consistent.
These are recorded points before the total is normalized to 100.
Public company/service identity; MCP identity only for MCP
These establish provenance, disclosure routes and consistency. They cannot certify prompt-injection resistance or safe transactions.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Could not read live MCP server identity from any declared endpoint (0 connected, 3 required authentication, 0 not probed before the scan deadline, 0 unreachable or invalid). Live MCP probe: 10748 ms, 3 protocol request(s).
Additional evidence for this capability. It earns no separate points, so observing another related signal does not count twice.
Public company/service identity; MCP identity only for MCP
These establish provenance, disclosure routes and consistency. They cannot certify prompt-injection resistance or safe transactions.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Instruction-safe interaction surfaces
Accessibility-tree probe unavailable (deep_scan_only): Accessibility-tree checks are deep-scan only to avoid always-on browser probe cost.
A limited scanner observation, not proof of access from real agent infrastructure. It remains unscored until the probe can support that claim.
Deep scans with an available browser accessibility tree snapshot
Prompt-injection-like instructions in accessibility text are a real manipulation risk for agent operators. This bounded probe can surface risky text but cannot prove complete absence, so it remains unscored.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Application isolation declarations
No adoption signal was detected, but endpoint discovery was incomplete or at least one declared MCP endpoint lacked readable list metadata, so site-level absence is unmeasurable. This metadata-only check did not call tools or read app resources.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected MCP application UI
Declared origins and CSP are relevant safeguards for an embedded UI, but declarations alone do not establish effective isolation.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
No adoption signal was detected, but endpoint discovery was incomplete or at least one declared MCP endpoint lacked readable list metadata, so site-level absence is unmeasurable. This metadata-only check did not call tools or read app resources.
An emerging convention, not a universal requirement. It adds no points and its absence subtracts none. Adoption and reliable testing must justify any future weight.
Detected MCP application UI
Declared origins and CSP are relevant safeguards for an embedded UI, but declarations alone do not establish effective isolation.
These sources inform the check. They do not validate our exact point weights.
Guidance recorded with this scan · 2026-09-09. Recorded scores are unchanged.
Agent Comprehension was unavailable for this scan.