Agent Readiness Score
replicate.com
45 — Developing. Agents can find parts of this site, but key capabilities are hard to reach.
HonestClaw measured how well replicate.com's site and APIs work for AI agents — across 9 dimensions and 45 automated checks.
Dimensions
Discoverability13.7/24
Content Accessibility11.9/24
Bot Access Control2/12
Protocol & Capability Discovery25.6/41
Authentication & Credentials0.5/19
Documentation Quality5.5/10
Structured Data I/O8/13
Observable State & Reliability3.7/15
Agent Safety & Trust4.7/11
Checks
Discoverability
Subdomain sweepsubdomain-sweepAn agent-relevant subdomain is live and serves its expected content kind (docs.replicate.com).Parseable
robots.txtrobots-txtrobots.txt at https://replicate.com/robots.txt declares a sitemap. Declared sitemap https://replicate.com/sitemap.xml resolves as valid XML.Cross-verified
API catalogapi-catalogAPI Catalog found at https://replicate.com/.well-known/api-catalog (application/linkset+json).Detected
sitemap.xmlsitemap-xmlhttps://replicate.com/sitemap.xml is a valid sitemap with lastmod and real (non-homepage) URLs.Useful
Canonical domaincanonical-domainRoot and www converge on replicate.com, with a matching canonical tag.Cross-verified
API existenceapi-existenceA real API surface is declared at https://api.replicate.com/openapi.json: the spec lists servers and paths. (source: spec)Parseable
API versioningapi-versioningAPI versioning is documented at https://replicate.com/docs/topics/models/versions (the versioning scheme). To reach the top level, it would need a documented deprecation/sunset lifecycle. (source: docs)Useful
Content Accessibility
llms.txtllms-txtA valid llms.txt header exists at https://replicate.com/llms.txt, but it lists few or no link sections (1 links, 7 sections).Parseable
Freshness signalsfreshness-signalsA single freshness signal is present (sitemap lastmod).Parseable
Crawl-cost estimatecrawl-cost-estimateInformational — not scoredSitemap index declares 6 child sitemap(s) — crawl cost is not directly observable from the index alone.Parseable
Raw content extractabilityraw-content-extractabilityReadable content is extractable without JavaScript on 2 probed surface(s) with substantial visible text.Cross-verified
llms-full.txtllms-full-txtNo llms-full.txt file found (HTML pages at .txt paths do not qualify).Not detected
ai.txtai-txtNo ai.txt file found (HTML pages at .txt paths do not qualify).Not detected
Markdown twinmarkdown-twinA markdown twin is published at https://replicate.com/docs.md alongside the docs HTML surface (https://replicate.com/docs).Parseable
Bot Access Control
AI bot rulesai-bot-rulesrobots.txt is readable but declares no AI-bot-specific rules.Not detected
Content signalscontent-signalsrobots.txt declares 3 recognized Content Signal(s) with a valid value: search, ai-train, ai-input.Cross-verified
RSL licensingrslrobots.txt is readable but declares no RSL License directive.Not detected
TDM reservationtdmrepNo TDMRep policy found at /.well-known/tdmrep.json.Not detected
Protocol & Capability Discovery
MCP descriptormcp-descriptorInformational — not scoredNo MCP descriptor found at the standard locations an agent would check.Not detected
MCP supportmcp-supportFirst-party documentation at https://replicate.com/docs/reference/mcp identifies the MCP endpoint https://mcp.replicate.com/ and explains both its capabilities and how clients connect or authenticate.Cross-verified
MCP Server Cardmcp-server-cardAn MCP Server Card at https://replicate.com/.well-known/mcp/server-card.json declares 1 remote endpoint(s) agents can connect to.Useful
A2A Agent Carda2a-agent-cardNo A2A Agent Card found at the standard well-known locations.Not detected
DNS-AIDdns-aidNo DNS-AID index found at _index._agents.replicate.com.Not detected
.well-known indexwell-known-indexNo non-owned /.well-known/ descriptors found.Not detected
SDK availabilitysdk-availabilitySDK libraries page at https://replicate.com/docs/reference/client-libraries: 9 GitHub repos, 0 registry links, 0 install commands (9 pointers).Cross-verified
GraphQL surfacegraphql-surfaceNo GraphQL transport surface found at conventional paths (GET-only; no introspection issued).Not detected
Auth discoveryauth-discoveryAgent-native auth discovery: a valid RFC 9728 PRM (via challenge) or a PRM plus a valid AS/OIDC descriptor.Cross-verified
Link headerslink-headersNo RFC 8288 Link headers found on any probed response.Not detected
OpenAPI specificationopenapi-specValid OpenAPI 3.1.0 spec at https://api.replicate.com/openapi.json: 36 operations, 19 schemas, 100% response coverage, servers declared.Cross-verified
Rate limit documentationrate-limit-docsRate limits are documented at https://replicate.com/docs/topics/predictions/rate-limits with a concrete limit and rate-limit response headers. To reach the top level, it would need 429 handling, Retry-After guidance, and tier/plan differences.Useful
WebMCPwebmcpInformational — not scoredWebMCP browser probing is not yet available in this scanner release.Couldn't verify
Authentication & Credentials
Credential managementcredential-managementNo credential management documentation found at the standard docs locations an agent would check.Not detected
Programmatic auth flowprogrammatic-auth-flowProgrammatic authentication is mentioned at https://replicate.com/docs, but not on a page dedicated to documenting it.Detected
Sandbox environmentsandbox-environmentNo sandbox or test environment documentation found at the standard docs locations an agent would check.Not detected
Auth documentationauth-documentationNo authentication documentation found at the standard docs locations an agent would check.Not detected
Documentation Quality
Changelog presencechangelog-presenceA changelog is published at https://replicate.com/changelog with multiple recent dated entries, indicating it is actively maintained.Cross-verified
API reference depthapi-reference-depthAPI reference content is mentioned at https://replicate.com/docs/reference/client-libraries, but not on a dedicated reference page.Detected
Structured Data I/O
Structured data I/Ostructured-data-ioThe spec at https://api.replicate.com/openapi.json declares typed, consistent response schemas (89% of operations) with a pagination pattern and a structured error envelope. (source: spec)Cross-verified
Machine-readable pricingmachine-readable-pricingNo machine-readable pricing endpoint found (HTML pricing pages do not qualify).Not detected
Data export APIdata-export-apiNo data export or bulk API documentation found at the standard docs locations an agent would check.Not detected
Observable State & Reliability
Status & health endpointsstatus-and-healthHealth/status endpoint found at https://www.replicatestatus.com/api/v2/status.json (HTTP 200).Detected
Retry & idempotencyretry-and-idempotencyNo retry or idempotency documentation found at the standard docs locations an agent would check.Not detected
Error documentationerror-documentationAn error documentation page exists at https://replicate.com/docs/reference/error-codes, but it doesn't enumerate specific error codes or a documented error response shape.Parseable
Webhook documentationwebhook-documentationA webhook documentation page exists at https://replicate.com/docs/topics/webhooks, but it doesn't list specific event types with their payload shape.Parseable
Machine payments (x402/MPP)x402-or-mpp-supportInformational — not scoredNo 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.Couldn't verify
Agent Safety & Trust
security.txtsecurity-txtNo security.txt found at /.well-known/security.txt or /security.txt.Not detected
Org identity signalsorg-identityHomepage exposes one org-identity signal (social).Detected
Metadata consistencymetadata-consistencyHomepage metadata (canonical, Open Graph, schema) is present and consistent.Cross-verified
Legal policy pageslegal-pagesBoth privacy and terms policies are discoverable with substantive content.Useful