reducto.ai agentic web readiness

A+ Manifest validity: 63% (5/8 valid, 8 errors, 17 warnings) · Surface score 65 (coverage 50 + breadth 15 + conformance 0). 8 standards observed across 5 of 9 steps; 8 manifests present (+ security.txt), 8 errors, 17 warnings. Checked 2026-10-25 03:17:28 UTC. This grades the agent-facing files published at reducto.ai, not the organisation's products.

Badge for a README: ![agentic web grade](https://agenticweb.wiki/badge/reducto.ai.svg) · JSON · Markdown · Re-check now

Coverage by step

Manifests

{
  "$schema": "https://static.modelcontextprotocol.io/schemas/v1/server-card.schema.json",
  "name": "ai.reducto/reducto",
  "title": "Reducto",
  "version": "1.0.0",
  "description": "Reducto and what it offers to agents",
  "websiteUrl": "https://reducto.ai/",
  "icons": [
    {
      "src": "https://reducto.ai/icon.png",
      "mimeType": "image/png",
      "sizes": [
        "180x180"
      ]
    }
  ],
  "remotes": [
    {
      "type": "streamable-http",
      "url": "https://reducto.ai/mcp",
      "supportedProtocolVersions": [
        "2025-11-25",
        "2025-06-18",
        "2025-03-26"
      ]
    }
  ]
}

{
  "name": "Reducto",
  "description": "Reducto and what it offers to agents",
  "version": "1.0.0",
  "provider": {
    "organization": "Reducto",
    "url": "https://reducto.ai/"
  },
  "documentationUrl": "https://reducto.ai/AGENTS.md",
  "supportedInterfaces": [
    {
      "url": "https://reducto.ai/a2a",
      "protocolBinding": "JSONRPC",
      "protocolVersion": "1.0"
    }
  ],
  "url": "https://reducto.ai/a2a",
  "protocolVersion": "1.0",
  "capabilities": {
    "streaming": false,
    "pushNotifications": false,
    "extendedAgentCard": false
  },
  "defaultInputModes": [
    "text/plain"
  ],
  "defaultOutputModes": [
    "text/plain"
  ],
  "skills": [
    {
      "id": "answer",
      "name": "Answer questions",
      "description": "Answer questions about Reducto.",
      "tags": [
        "qa"
      ],
      "examples": [
        "What does Reducto do?"
      ]
    }
  ],
  "securitySchemes": {},
  "securityRequirements": []
}

# Authenticating to Reducto as an agent

This file is the agent-auth skill for reducto.ai. It is linked from the authorization server's metadata (`agent_auth.skill`).

## Discovery
1. Call the API. A `401` carries `WWW-Authenticate: Bearer resource_metadata="https://reducto.ai/.well-known/oauth-protected-resource"`.
2. Read `/.well-known/oauth-protected-resource` (RFC 9728) for `resource` and `authorization_servers`.
3. Read the authorization server's `/.well-known/oauth-authorization-server` (RFC 8414); the `agent_auth` block lists the endpoints below.

## Methods we accept
- `identity_assertion`: you hold a session for a user and can mint an ID-JAG with `aud` = our `resource`. Preferred.
- `service_auth`: you only know the user's email. A claim ceremony (6-digit code) is required.
- `anonymous`: no user. Immediate limited scopes; the user can claim later.

## Register
POST `agent_auth.identity_endpoint` with `{"type": "<method>", ...}`; exchange the returned `identity_assertion` at the token endpoint for an access token.

## Scopes
- `api.read`: read
- `api.write`: write (claimed registrations only)

## Rules
- Show the user our name and the scopes before asserting their identity.
- Keep `claim_token` in memory only.
- Revocation events arrive as Security Event Tokens at your registered events endpoint.

What an agent can call

• MCP endpoint https://mcp.reducto.ai/mcp — from /.well-known/mcp.json
• MCP endpoint https://mcp.reducto.ai/mcp (http), tools: get_documentation, upload_file, parse_document, extract_data, split_document, classify_document, edit_document, get_job, list_jobs — from /.well-known/mcp/server-card.json
• MCP endpoint https://docs.reducto.ai/mcp (http) — from /.well-known/mcp/docs-server-card.json
• A2A agent Reducto at https://mcp.reducto.ai/mcp, skills: Parse, Extract, Split, Classify, Edit
• A2A agent Reducto at https://mcp.reducto.ai/mcp, skills: Parse, Extract, Split, Classify, Edit
• API description https://accounts.reducto.ai/api/agent/openapi.json (application/json)
• API description https://reducto.ai/auth.md (text/markdown)
• API description https://reducto.ai/openapi.json (application/json)
• API description https://docs.reducto.ai/openapi.json (application/json)
• API description https://reducto.ai/.well-known/agent-card.json (application/a2a-agent-card+json)
• API description https://reducto.ai/.well-known/agents.json (application/json)
• API description https://reducto.ai/.well-known/mcp.json (application/json)
• API description https://reducto.ai/.well-known/webmcp.json (application/json)
• API description https://reducto.ai/.well-known/mcp/server-cards.json (application/json)
• API description https://reducto.ai/openapi.json (application/json)
• API description https://docs.reducto.ai/openapi.json (application/json)
• API description https://docs.reducto.ai (text/html)
• API description https://reducto.ai/.well-known/ai-catalog.json (application/ai-catalog+json)
• API description https://docs.reducto.ai/mcp-server (text/html)
• API description https://reducto.ai/.well-known/mcp/server-card.json (application/mcp-server-card+json)
• llms.txt: Reducto, 58 links, llms-full.txt present
• robots.txt AI policy: unlisted crawlers partial
• ARD bridge (/.well-known/ai-catalog.json): 1/7 valid, 3 legacy, 1 dead/broken (served at deprecated path only)

Standards observed

• Agent2Agent Protocol https://reducto.ai/.well-known/agent-card.json
• llms.txt https://reducto.ai/llms.txt
• Content Signals Policy https://reducto.ai/robots.txt
• AI Catalog well-known file https://reducto.ai/.well-known/ai-catalog.json
• api-catalog: A Well-Known URI and Link Relation to Help Discovery of APIs https://reducto.ai/.well-known/api-catalog
• AI Crawler User-Agent Tokens in robots.txt https://reducto.ai/robots.txt
• Agent Skills https://reducto.ai/.well-known/agent-skills/index.json
• Markdown for Agents https://reducto.ai/

Cheapest next steps

1. Agentic Resource Discovery: a catalog at /.well-known/ard.json lets agents discover your APIs, MCP servers and data in one fetch (emerging spec — publish only if you run the underlying endpoints; generate via MCP template({"id": "ard", "domain": "reducto.ai"}))
2. AGENTS.md: instructions for coding agents working on or with the site
   Generate starter /AGENTS.md for reducto.ai

   # Reducto for agents
   
   Reducto and what it offers to agents
   
   ## Read the data, not the HTML
   - Index: https://reducto.ai/llms.txt
   - API description: https://reducto.ai/openapi.json
   - Discovery catalog: https://reducto.ai/.well-known/ard.json
   
   ## Calling
   - MCP server (Streamable HTTP): https://reducto.ai/mcp
   - Rate limits and authentication: describe them here
   
   ## Conventions
   - How to cite or attribute content
   - What not to do (no scraping of /account/, no automated sign-ups, etc.)
   
   Contact: security@reducto.ai
   

Agent pre-flight & practical readiness

Preferred mode: mcp (fallback openapi). Call the MCP endpoint at https://mcp.reducto.ai/mcp first. Do not scrape the HTML when a tool can answer.

Fetch: POST https://mcp.reducto.ai/mcp POST a JSON-RPC tools/list, or server/discover on the 2026-07-28 revision, before reading HTML.

Practical readability: exa_parsable yes, markdown negotiation yes, render markdown_native, HTML quality high, first content at byte 0.

AI crawler posture: open_to_all · Assistant access: OpenAI partial, Anthropic partial, Google partial, Perplexity partial, Exa allowed.

Readiness subscores: reach 20/20, readability 30/30, capabilities 35/35, conformance 0/15, total 85/100, grade A+.

How the grade is made

+10 per lifecycle step with a standard observed, counting manifests that validate or that the probe rule for the standard recognises (a valid MCP server card counts for MCP; max 90); +3 per observed standard beyond the first two (max 15); conformance +15 (+10 with one manifest) minus 5 per agent-facing manifest with errors, never below 0 (security.txt is reported, not scored). Lifecycle grade: A+ >= 60, A >= 45, B >= 30, C >= 20, D >= 10. Strict 100-point composite (subscores.total_100 / subscores.grade_100): reach_and_policy (20) + readability (30) + capabilities (35) + conformance (15), where A+ >= 80, A >= 65, B >= 45, C >= 28, D >= 15. Errors are specification violations; warnings are best-practice gaps. Only public paths are fetched, without credentials, so standards that need a login (skipped in the probe) are not counted either way. The same report is the validate_manifest tool on the MCP server.

JSON · Markdown · probe rules