MCP Server
The entire API Evangelist network is available to AI agents as a Model Context Protocol server over Streamable HTTP:
Code
It covers the same resources as the REST API, plus concierge tools that bundle curated research for a topic. The server card is published at https://apievangelist.com/.well-known/mcp/server-card.json.
Start with the concierge
Two tools are designed as the entry point — agents should reach for these first:
get_overview— start here: what the network contains, counts per resource type, and how the tools fit together.guide_topic(topic)— a curated bundle for one topic: the area plus its guidance, rules, policies, papers, stories, and services in a single call.
Tools
| Tool | What it does |
|---|---|
get_overview | Start here — network contents, counts, and how the tools fit together |
guide_topic(topic) | Curated bundle: area + guidance/rules/policies + papers + stories + services |
find_services / get_service | API Evangelist services — Kin's own offerings, the front door to an engagement |
find_solutions / get_solution | The solutions registry — 740+ third-party API services and tools ranked by adoption |
search_api_evangelist | Unified search across every resource type, faceted with types= |
find_posts / get_post | 5,100+ blog stories since 2010 |
find_areas / get_area | The 77 topic areas and their related-area graph |
find_building_blocks(type) / get_building_block(type, slug) | Guidance, rules, policies, standards, strategies, schema, properties, experiences, lifecycle |
get_guidance_section(section) | One editorial guidance section — history, technology, business, politics, governance, or evangelism |
find_tools / get_tool | The open-source tooling registry — 400+ tools ranked by adoption, with radar placement |
find_conversations / get_conversation | Conversations with API producers, consumers, and vendors |
find_videos / get_video | The video library, searchable by transcript |
find_papers / get_paper | White papers |
get_timeline(timeline) | The api, resources, industries, and people research timelines |
find_vocabulary(category) | The API vocabulary |
get_feeds | Machine-readable index of the static JSON feeds |
find_company_insights · insights_dimensions · insights_adoption | Insights (demand-side) — what ~5,800 companies build, buy, and hire for. Free discovery. |
get_company_insight · company_gaps | Insights depth — a company's full 40-dimension profile and its gaps. Pro. (The provider-match join lives on the apis.io MCP.) |
get_contact | Every way to reach API Evangelist |
Governance & Discovery tools
The Governance & Discovery API is on the MCP server too — one tool per
REST operation, so anything you can curl an agent can call. Same engine, same rules, same
line: checking an API is free and keyless; generating an artifact, certifying, auditing a
repo, or rolling up across an estate is Pro.
| Tool | Tier | What it does |
|---|---|---|
validate_api | Free | Lint OpenAPI / AsyncAPI / Arazzo / JSON Schema against the curated ruleset — or pass your own ruleset. |
api_coverage | Free | How much of a description your governance actually inspects. |
api_scorecard | Free | 0–100 maturity score + letter grade across design, coverage, docs, and agent-readiness. |
diff_api_versions | Free | Breaking-change detection between two OpenAPI versions. |
classify_api_fields | Free | Inventory the PII and secret fields an API exposes. |
deprecation_posture | Free | Which deprecated operations actually announce a sunset. |
reconcile_waivers | Free | Waivers vs lint output → the effective failing set to gate on. |
governance_report · verify_certificate | Free | HTML governance report; re-verify a certificate against its API. |
validate_onboarding | Free | Validate an API Onboarding Descriptor (AID). |
certify_api | Pro | Issue a tamper-evident governance certificate. |
export_agent_rules | Pro | Ruleset → AGENTS.md, system prompt, remediation pack, digest. |
audit_pipeline | Pro | Score a repo's Spectral CI setup + punch-list. |
forge_mcp_tools | Pro | Forge MCP tools + Agent Skills per operation from an OpenAPI. |
gate_agent_context | Pro | Emit a governed Tyk + MCP + ruleset agent surface. |
apply_overlay | Pro | Apply an OpenAPI Overlay and report what each action matched. |
mock_payloads · code_snippets | Pro | Example payloads; runnable curl / JavaScript / Python per operation. |
score_reusability | Pro | Reuse + duplication across a set of APIs. |
map_api_experience | Pro | The API → MCP → Agent-Skill chain, with coverage. |
generate_agent_descriptor | Pro | An API's llms.txt + AGENTS.md consumption contract. |
Two Agent Skills describe this surface for agents:
api-governance (check,
gate, certify) and
api-agent-readiness
(make an API consumable by agents).
A note on rulesets. The default catalog is a best-of-breed superset — run it raw against
a real API and you get thousands of warnings and no errors, which is how teams learn to ignore
governance. Pass a ruleset you own to validate_api, api_coverage, and api_scorecard and
gate on that instead.
Prompts
Prompts are guided flows — each chains the right tools for a job, so an agent (or you) doesn't have
to know the tool names. Free prompts run for anyone; Pro prompts are listed for everyone and
upsell on use. Call prompts/list for the always-current set, prompts/get to run one.
Research & learn — where_can_ae_help, catch_up, learn_a_topic, weekly_digest,
find_the_paper, explore_topic_area, validate_onboarding_descriptor; research_topic Pro,
vendor_landscape Pro
Check an API (free) — governance_baseline, classify_exposure, version_change_review,
waivers_review, governance_html_report
Make an API agent-ready — agent_ready_audit (free first pass); make_my_api_agent_ready
Pro, forge_mcp_server Pro, generate_consumption_contract Pro, mock_and_snippets Pro
Govern & prove at scale — govern_my_estate Pro, make_internal_apis_agent_safe Pro,
pipeline_audit Pro, certify_and_prove Pro, apply_governance_overlay Pro,
reusability_review Pro, map_experience Pro, governance_starter_kit Pro
Demand-side — company_demand_profile (free preview / Pro full); where_to_sell Pro
Code
Resources
Attachable, readable context an agent can pin into a session. Free unless marked Pro.
ae://overview,ae://services,ae://contact— the network root, Kin's services, and how to reach outae://statsPro — per-type counts, stories-per-year, top tags- Templated:
ae://area/{slug},ae://tag/{slug},ae://post/{slug}(free);ae://guide/{topic}Pro
Call resources/list and resources/templates/list for the authoritative set. The
experience.apicommons.org tool renders any APIs.json as a
visual REST → MCP → Agent-Skill map — the same chain map_experience reports.
Connect
Claude Code
Code
Claude Desktop / .mcp.json
Code
Cursor (~/.cursor/mcp.json)
Code
VS Code (.vscode/mcp.json)
Code
Poke it directly
The server speaks JSON-RPC — initialize, tools/list / tools/call, prompts/list / prompts/get, and resources/list / resources/read:
Code
The same tiers and limits apply as the REST API — keyless free tier, with higher limits by request.