ambiguous.ai
Finalizing the report
Preparing the stored Is Agentic score
Essential—
Recommended—
Bonus signals—
ambiguous.ai
Preparing the stored Is Agentic score
ambiguous.ai
These checks describe whether an ordinary agent can enter, read, and operate the public site.
Agents can reach the site
Crawler access and bot defenses.
Core content is available
Useful content remains accessible without a fragile browser-only path.
Navigation fails safely
Redirects and missing pages give agents a recoverable path.
Controls are understandable
Forms and interactive controls expose usable names and structure.
The public website is always evaluated. Optional surfaces appear when the scan finds positive evidence that they apply.
Strong
Weighted across 16 applicable checks
Strong
Weighted across 12 applicable checks
Strong
Weighted across 3 applicable checks
Ready with gaps
Weighted across 5 applicable checks
Strong
Weighted across 1 applicable check
Critical access gaps come first, followed by other applicable readiness gaps.
No material issues were found in the applicable checks.
Checks are grouped by relevance for App sites.
9853 chars, semantic headings (1 H1 + 7 H2s), 5.7% content ratio
Site accessible to 6 AI agent user-agents
No meta-refresh stubs, JavaScript-redirect stubs, or cross-domain hops across 6 checked pages
All 5 sampled pages are publicly readable (5 with substantive content)
OpenAPI spec found at https://app.ambiguous.ai/api/openapi.json (version: 3.1.0)
Canonical URL serves text/markdown and text/html via Accept negotiation with Vary: Accept
OpenID Connect discovery endpoint found at https://app.ambiguous.ai
Zero-auth service - no scoping needed, all endpoints are openly accessible to agents
API returns JSON error responses (401 at https://app.ambiguous.ai/sandbox (from OpenAPI servers))
MCP server exposes 2 resource(s) via resources/list
Nonexistent paths return HTTP 404 with markdown guidance for agents - the strongest 404 contract
Valid sitemap found at https://www.ambiguous.ai/sitemap.xml with 41 entries
Rich JSON-LD identity: SoftwareApplication with name, description, url, and category/offers (4 blocks)
Pricing page found at /pricing
API/docs link found on homepage and resolves: /applications/docs
All metadata signals present: canonical URL, lang="en", og:image, og:type
Organization schema complete with contactPoint and address
All trust anchor pages verified: About, Contact, Privacy
All 9 measured pages fit an agent context budget (largest ~2K tokens)
REST API documentation found at https://www.ambiguous.ai/.well-known/api-catalog. Best-of-protocols score: 7/7.
Low friction onboarding verified live: free tier available, self-serve key generation, sandbox/test environment, zero-auth access, zero-friction self-serve registration at /register
MCP server connected via Streamable HTTP - ambiguous-workspace v0.1.0 (protocol 2025-11-25)
The live API at https://ambiguous.ai/api returns IETF RateLimit-* headers: ratelimit-policy.
OpenAPI defines a typed error schema in components.schemas and 4xx/5xx responses reference it
API versioning strategy found (header versioning) with sunset/deprecation markers documented
CLI package found on npm: ambiguous
94% of operations define typed response schemas, 94% use application/json
REST: agent-friendly schema (1427 operations, 100% with operationIds, 100% documented by description or summary)
Compatible: 1427/1427 ops with IDs, 1417/1427 with typed schemas
2/2 resources read with valid mimeType and non-empty content
ARD catalog valid at /.well-known/ard.json (spec 1.0) - 4/4 entries, trustManifest present
Valid AI Catalog at /.well-known/ai-catalog.json (spec 1.0) - 4/4 entries
All 4 entries valid with domain-anchored urn:air identifiers (catalog at /.well-known/ard.json)
trustManifest present on catalog entries (4/4 entries valid, catalog at /.well-known/ard.json)
Agent discovered 1 developer-resource type by name (API docs) across 1 page
ambiguous.ai ranks #4 for "Ambiguous AI workspace"
NPM package found: ambiguous - "CLI for Ambiguous Workspace — manage docs, tasks, mail, calendar, CRM, and more from the terminal"
MCP manifest has full branding (name, icon, description) at /.well-known/mcp/server-card.json
Found in ChatGPT app directory: "Ambiguous Workspace"
Agent Skills index (agentskills.io) found at /.well-known/agent-skills/index.json with 1 skill(s)
Index v0.2.0 conformant; verified SHA-256 of "ambiguous" (skill-md)
A2A Agent Card found: "Ambiguous Workspace"
pricing.md found at /pricing.md but thin (103 lines) - add plan tiers, prices, and feature breakdowns
How to pass
Create a /pricing.md file with your pricing tiers, features, and limits in plain markdown. This lets AI agents compare costs and recommend plans without scraping HTML pricing pages.
Valid Schema Map XML found at https://www.ambiguous.ai/schemamap.xml
MCP server discoverable via server-card.json at /.well-known/mcp/
Rich agent mode view at ?mode=agent with 6 agent signals and navigation links
RFC 8288 Link header advertises: sitemap, describedby, service-desc, api-catalog, alternate(markdown)
Homepage markdown fallback works (https://ambiguous.ai/index.md). No additional same-host content pages were discovered to sample.
Modular llms.txt files found for sections: applications, agents
98% of 41 sampled sitemap entries carry lastmod; newest is 1 day(s) old
Found the llms.txt at https://ambiguous.ai/llms.txt.
The llms.txt is well-formatted: 63 lines with markdown links, 6,013 characters in total.
When-to-use guidance found in llms.txt
Strong entity linking via sameAs: linkedin.com, github.com
Some extended schema types found: Service, FAQPage - add FAQPage, Service, or AggregateRating for full coverage
How to pass
Expand your JSON-LD beyond Organization/WebSite. Add FAQPage for common questions, Service or Product for offerings, AggregateRating or Review for social proof, and BreadcrumbList for navigation context.
All 5 probed llms.txt links resolve to real content
Markdown alternate advertised and verified: https://www.ambiguous.ai/index.md serves markdown
Frontmatter with title + description and canonical on /index.md
Code fences balanced across 1 markdown document
Developer portal found at /developers
RFC 9727 api-catalog with 1 API link(s) and rfc9727 profile
Path-suffix markdown docs served with text/markdown content-type: /agents.md, /auth.md, /index.md
https://ambiguous.ai serves markdown to the GPTBot User-Agent even with Accept: text/html - UA-sniffed agent responses in place
All 877 tools on product MCP have detailed descriptions (>= 20 chars)
797/877 tools on product MCP have parameter schemas, 678 with required fields
product MCP identifies as "ambiguous-workspace" v0.1.0 with instructions
product MCP exposes 877 tools - substantial tool surface
All 877 tool names follow consistent convention, descriptive, and non-generic
docs MCP is public - correct posture for documentation surface
Web Bot Auth directory has no 'keys' array
How to pass
Publish a Web Bot Auth directory at /.well-known/http-message-signatures-directory. Serve a JSON document with a 'keys' array of Ed25519 JWKs (kty=OKP, crv=Ed25519, kid, nbf, exp). This lets agents sign their requests per RFC 9421 so you can distinguish legitimate bots from spoofers.
RFC 9728 metadata at https://app.ambiguous.ai: resource=https://app.ambiguous.ai/mcp, authorization_servers (1), scopes_supported (1), bearer_methods_supported
auth.md at https://ambiguous.ai/auth.md served as text/markdown; charset=utf-8 (5000 chars)
WorkOS auth.md structure at https://ambiguous.ai/auth.md - 6/8 walkthrough sections (found Discover, Pick a method, Register, Claim, Errors, and Revocation); anchors: oauth-protected-resource, oauth-authorization-server, and agent_auth; register shapes: service_auth shape
Static fallback (deep check disabled): PRM + AS metadata both reachable with agent_auth.identity_endpoint (https://app.ambiguous.ai/oauth/agent/identity); walkthrough discoverable end-to-end
PRM + AS metadata cross-linked (PRM discovered at https://app.ambiguous.ai) (AS metadata fetched from advertised origin https://app.ambiguous.ai); agent_auth has identity_endpoint (https://app.ambiguous.ai/oauth/agent/identity), identity_types_supported [service_auth] with the required per-type blocks, and skill -> https://ambiguous.ai/auth.md (also advertised: claim_endpoint; events_endpoint, events_supported absent)
401 at https://app.ambiguous.ai at /mcp with spec-shaped WWW-Authenticate hint: Bearer resource_metadata="https://app.ambiguous.ai/.well-known/oauth-protected-resource"
agent_auth endpoints reachable (source: AS metadata agent_auth block) - identity_endpoint: HTTP 204, claim_endpoint: HTTP 204, events_endpoint: not advertised
product MCP returns structured JSON-RPC errors with code and message
product MCP uses modern Streamable HTTP transport
Imperative WebMCP API detected (document.modelContext.registerTool in inline script).
REST Idempotency-Key header found on POST /api/mail/send in OpenAPI spec
Cursor-based pagination found in OpenAPI spec response schemas or query parameters
Async job pattern detected: 202 Accepted with a Location header or job-tracking fields on the 202 body
SDK package found only in npm
How to pass
Publish official SDK packages across multiple language ecosystems (npm, PyPI, Go modules, RubyGems). Auto-generate them from your OpenAPI spec using tools like openapi-generator. For each package set the project URL or homepage to your product domain (package.json repository/homepage, PyPI Home-Page or project_urls, RubyGems homepage_uri) - this is how agents verify the package is your official SDK.
NLWeb /ask supports SSE streaming with valid event data (via GET)
product MCP: 877/877 tools have behavioral annotations (readOnlyHint/destructiveHint)
MCP server card found at https://www.ambiguous.ai/.well-known/mcp/server-card.json (3 tools advertised) (note: card leaves out 874 live tools: users_list, auth_whoami, get_users_notification_preferences, ...)
You run both: a product MCP server (https://app.ambiguous.ai/mcp) for actions and a docs MCP server (https://www.ambiguous.ai/docs/mcp) for documentation.
Batch endpoint found in OpenAPI spec (/batch path)
Server HTML is a well-structured document (main=true, landmarks=3/4, h1=1, maxHeadingSkip=1).
147 native controls, 10 non-native div-soup affordances (94% native).
147/147 interactive elements have a computable accessible name (100%).
No hidden instruction text detected in accessibility-tree attributes or off-screen content.
MCP Apps: 2 ui:// resources, 9 UI-enabled tools (create_doc, get_document, create_sheet), correct MIME type
MCP Apps: 2 interactive ui:// resource(s) for in-conversation UI rendering
Good: correct MIME type, valid HTML5. Missing: no color-scheme meta tag
How to pass
Ensure your MCP Apps resources use MIME type text/html;profile=mcp-app, include , and add <meta name="color-scheme" content="light dark"> for dark mode. Never hardcode secrets in resource HTML.
Inline ui:// view delivered with valid HTML and no auth wall
ACP checkout endpoint reachable at https://api.ambiguous.ai/checkout_sessions (OPTIONS 204 allows POST, POST 200 confirmed)
Delegate Payment endpoint reachable at https://api.ambiguous.ai/agentic_commerce/delegate_payment (OPTIONS 204 allows POST, POST 200 confirmed)
What does ambiguous.ai do and who is it for? Explain it back to me.
Sign me up for Vercel product updates and marketing emails.
Unsubscribe anytime. Privacy Notice
Source: Ora API
What does ambiguous.ai do and who is it for? Explain it back to me.
8 steps2 reasoning steps1 search
The agent successfully assembled a coherent explanation of Ambiguous.ai by fetching the homepage and pricing page directly, supplemented by third-party positioning from a16z. However, the site's own navigation is sparse—several expected routes (how-it-works, features) returned 404s—forcing the agent to type URLs from memory and rely on external sources for differentiation details.