# Works With Agents — Agent Discovery Index > Open specifications, SDK references, and live discovery endpoints for AI agent interoperability. > **Status:** Live on `https://workswithagents.dev`. > **Full reference:** https://workswithagents.dev/llms-full.txt > **OpenAPI:** https://workswithagents.dev/v1/openapi.json > **Sitemap:** https://workswithagents.dev/sitemap.xml ## Agent entry points Base URL: `https://workswithagents.dev` Version prefix: `/v1` ```http GET /llms.txt # this concise agent index GET /llms-full.txt # fuller agent-readable reference GET /v1/openapi.json # OpenAPI 3.1 contract GET /v1/health # API health GET /v1/version # sync/version hash GET /v1/counts # public live counts for facts, pitfalls, specs, repos, agents GET /specs/ # specification directory GET /specs/index.md # published spec index GET /mcp # MCP server information GET /benchmarks/ # LLM coding + MCP quality benchmarks ``` ## Live API endpoints ```http GET /v1/facts?entity=X&category=Y&limit=200 POST /v1/facts GET /v1/facts/stats GET /v1/facts/verified GET /v1/skills GET /v1/skills/{name} GET /v1/pitfalls?tool=X&keyword=Y&limit=100 POST /v1/pitfalls GET /v1/pitfalls/stats GET /v1/stats?days=7 GET /v1/benchmarks POST /v1/newsletter/subscribe POST /v1/contact GET /v1/site-config ``` `GET /v1/auth/{service}` exists for authorized local agent infrastructure. Do not assume public access. ## MCP Server — Agent-native tools Hosted endpoint: `https://workswithagents.dev/mcp` Full package: `pip install wwa-mcp` 14 MCP tools exposing agent infrastructure: facts, pitfalls, skills, blueprints, handoff protocol, spec search, trust-score validation, capability discovery, deployment generation, SLAs, compliance checks, identity verification, onboarding generation, reputation queries. Configure in any MCP client with `"url": "https://workswithagents.dev/mcp"` (HTTP JSON-RPC transport). ## Published specifications — 23 specs (CC BY 4.0) Framework: - https://workswithagents.dev/specs/agent-osi-model.md Cross-framework: - https://workswithagents.dev/specs/asfs.md - https://workswithagents.dev/specs/agent-coding-benchmark.md Layer 1 — Entry: - https://workswithagents.dev/specs/onboarding.md Layer 2-3 — Identity & Capability: - https://workswithagents.dev/specs/identity.md - https://workswithagents.dev/specs/capability-manifest.md - https://workswithagents.dev/specs/trust-score.md - https://workswithagents.dev/specs/reputation-ledger.md Layer 4 — Session: - https://workswithagents.dev/specs/handoff.md - https://workswithagents.dev/specs/clarification-protocol.md Layer 5 — Coordination: - https://workswithagents.dev/specs/coordination.md - https://workswithagents.dev/specs/iacp.md - https://workswithagents.dev/specs/iacp-internet-draft.md (IETF RFC format) Layer 6 — Economics & Security: - https://workswithagents.dev/specs/agent-economics.md - https://workswithagents.dev/specs/security-disclosure-protocol.md Layer 7 — Audit & Governance: - https://workswithagents.dev/specs/transaction.md - https://workswithagents.dev/specs/sla-framework.md - https://workswithagents.dev/specs/compliance-as-code.md - https://workswithagents.dev/specs/attestation-protocol.md - https://workswithagents.dev/specs/auditor-verification.md - https://workswithagents.dev/specs/deployment-manifest.md - https://workswithagents.dev/specs/local-first.md - https://workswithagents.dev/specs/ecp.md Published index: https://workswithagents.dev/specs/index.md ## SDKs and source - Python: `pip install workswithagents` - TypeScript: `npm install @workswithagents/agent-foundry` - Source: https://github.com/workswithagents/works-with-agents ## Methodology — How to work with AI agents The 13-module methodology for working effectively with AI agents. Free, no paywall. ```http GET /learn # 13 modules grouped by phase (Foundation, Autonomy, Scale, Harden, Oversight) GET /learn/01-boot # Pattern 1: First session setup — AGENTS.md, environment, context GET /learn/02-skills # Pattern 2: Reusable procedural knowledge — build once, use forever GET /learn/03-memory # Pattern 3: Durable context — never re-explain yourself GET /learn/04-decision-protocols # Pattern 4: Autonomy without chaos GET /learn/05-tool-composition # Pattern 5: Right tool for each job GET /learn/06-orchestration # Pattern 6: Multi-agent workflows GET /learn/07-pipelines # Pattern 7: Agents that run while you sleep GET /learn/08-resilience # Pattern 8: Never-stop loops GET /learn/09-verify # Pattern 9: Trust but verify GET /learn/10-compounding # Pattern 10: The feedback loop ``` See https://workswithagents.dev/learn for the full index. ## FAQ — Common AI agent questions Structured Q&A with JSON-LD FAQ schema. Answers to the most common questions about AI agents. ```http GET /faq # 15 questions covering methodology, memory, skills, orchestration, infrastructure ``` Questions answered include: how to start with AI agents, agent memory, skills, multi-agent orchestration, autonomous pipelines, agent communication, llms.txt, output verification, error handling, agent vs chatbot, tool selection, self-improvement, model selection, infrastructure setup, and the Agent OSI Model. Each answer draws from real experience running 25+ autonomous agents. ## Blog ```http GET /blog # All posts — infrastructure, agents, and what breaks in production GET /blog/{slug} # Individual post (rendered from markdown) ``` ## Crawl guidance - Prefer `llms-full.txt` for full context. - Prefer `v1/openapi.json` for endpoint schemas. - Prefer `specs/index.md` for the published spec list. - Send an explicit User-Agent such as `wwa-mcp/1.0` or `workswithagents-agent/1.0` when possible. Generic `Python-urllib/*` is allowed on the public machine-readable endpoints listed here. - Do not scrape private credentials or assume `/v1/auth/{service}` is public.