com.apievangelist/api-evangelist

Measured conclusion: Alive, not deep-tested. L2 outcome: not_attempted.

Sixteen years of API research as MCP tools — stories, areas, governance blocks & services

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "api-evangelist": {
      "url": "https://mcp.apievangelist.com/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (68)

• get_overview read-only

  START HERE. What the API Evangelist network offers: sixteen years of API research as searchable stories, 77 topic areas, governance building blocks (guidance, rules, policies, standards, strategies), papers, conversations, and the consulting services Kin Lane offers teams. Returns counts, entry points, and how to engage.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• guide_topic read-only

  PRO — Curated guide to one API topic (e.g. "gateways", "api governance", "discovery", "agents"). Bundles the matching topic area (with related areas), top guidance, rules, policies, and standards, relevant papers and recent stories, plus the API Evangelist services that can help a team with this topic.

  Parameters:

  • topic (required) — string — The topic to build a guide for.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_services read-only

  The services API Evangelist offers teams and customers — API strategy, governance, discovery, and evangelism engagements by Kin Lane. The front door to working with API Evangelist. (For the third-party vendor registry, use find_solutions.)

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_service read-only

  Full detail for one service offering by slug; include=["content"] inlines the full pitch.

  Parameters:

  • slug (required) — string — Service slug, as find_services returns it, e.g. "governance-agent-readiness".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • include — array — Pass ["content"] to inline the full body (Pro). Omitted, you get the summary.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_solutions read-only

  The solutions registry — 740+ third-party API services and tools (gateways, portals, testing, monitoring, …) ranked by adoption across the companies API Evangelist tracks.

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_solution read-only

  One solution from the vendor registry by slug, with its website and adoption count.

  Parameters:

  • slug (required) — string — Solution (vendor) slug, as find_solutions returns it, e.g. "adobe".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_tools read-only

  The open-source tooling registry — 400+ tools in use across the companies API Evangelist tracks, each with adoption counts and technology-radar placement.

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_tool read-only

  One tool from the tooling registry by slug, with its website, adoption count, and radar ring.

  Parameters:

  • slug (required) — string — Tool slug, as find_tools returns it, e.g. "ajv".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_contact read-only

  Every way to reach API Evangelist — email, contact form, newsletter, GitHub, social channels, and postal mail.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• search_api_evangelist read-only

  Unified search across the whole network: 5k+ stories, 77 topic areas, 2.5k governance building blocks, papers, conversations, videos, tools, timelines, services, and vocabulary. Filter with types (posts, areas, guidance, rules, policies, standards, strategies, schema, properties, experiences, lifecycle, conversations, videos, papers, services, solutions, tools, timelines, vocabulary).

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • types — array — Content types to include — plural resource names: posts, services, solutions, tools, areas, papers, videos, conversations, guidance, rules, policies, standards, strategies…, e.g. ["posts","papers"].
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_posts read-only

  Search or browse 5,100+ API Evangelist stories (2010–present); filter by year or tags.

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • year — string — Four-digit year; only posts whose date starts with it, e.g. "2024".
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_post read-only

  One story by slug (YYYY-MM-DD-slug); include=["content"] returns the full markdown body.

  Parameters:

  • slug (required) — string — Post slug — date-prefixed, as find_posts returns it, e.g. "2026-09-29-you-cannot-see-your-own-api-from-the-outside".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • include — array — Pass ["content"] to inline the full body (Pro). Omitted, you get the summary.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_areas read-only

  The 77 focused topic areas of the network (agents, gateways, security, …), each a subsite with its own catalog.

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_area read-only

  One topic area with its related-area graph resolved — walk the network from here.

  Parameters:

  • slug (required) — string — Area slug, as find_areas returns it, e.g. "agents".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_building_blocks read-only

  Search the governance building blocks by type: guidance (177), rules (617), policies (283), standards (432), strategies (75), schema (11), properties (790), experiences (46), lifecycle (29).

  Parameters:

  • type (required) — string — Building-block type: guidance, rules, policies, standards, strategies, schema, properties, experiences or lifecycle.
  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_building_block read-only

  One building block by type + slug; include=["content"] returns the full body.

  Parameters:

  • type (required) — string — Building-block type: guidance, rules, policies, standards, strategies, schema, properties, experiences or lifecycle.
  • slug (required) — string — Building-block slug within `type`, as find_building_blocks returns it, e.g. "access" with type "guidance".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • include — array — Pass ["content"] to inline the full body (Pro). Omitted, you get the summary.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_conversations read-only

  API Evangelist Conversations — recorded discussions with API practitioners (guest, company, YouTube/SoundCloud links).

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_conversation read-only

  One conversation with guest detail and media links; include=["content"] returns the writeup.

  Parameters:

  • slug (required) — string — Conversation slug — date-prefixed, as find_conversations returns it, e.g. "2026-09-01-lukasz-gornicki".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • include — array — Pass ["content"] to inline the full body (Pro). Omitted, you get the summary.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_papers read-only

  API Evangelist papers — single-PDF deep dives on one practice or topic, with outlines.

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_paper read-only

  One paper with its full outline and purchase/landing URL.

  Parameters:

  • slug (required) — string — Paper slug, as find_papers returns it, e.g. "change-on-the-agent-surface".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_videos read-only

  The API Evangelist video library — interviews and discussions from across the API space, searchable by transcript.

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_video read-only

  One video by slug with its YouTube id; include=["content"] returns the full transcript.

  Parameters:

  • slug (required) — string — Video slug, as find_videos returns it, e.g. "adam-duvander".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • include — array — Pass ["content"] to inline the full body (Pro). Omitted, you get the summary.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_timeline read-only

  One of the four research timelines in chronological order: api (the moments that shaped web APIs), resources (which digital resources became APIs and when), industries (the industrial history framing API infrastructure), people (the people who made compute and APIs possible).

  Parameters:

  • timeline (required) — string — Which timeline: api, resources, industries or people.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_guidance_section read-only

  One editorial guidance section as a list: history, technology, business, politics, governance, or evangelism — the six lenses API Evangelist organizes guidance through.

  Parameters:

  • section (required) — string — Guidance section: history, technology, business, politics, governance or evangelism.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_vocabulary read-only

  The API vocabulary — thousands of resources, actions, personas, domains, schemas, and tags extracted from the API landscape. Filter by category.

  Parameters:

  • q — string — Free text over title, summary, tags, and body.
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — Order when there is no `q`: "default" (dated content newest first, the rest A-Z), "title" (A-Z) or "date" (newest first). A `q` always orders by relevance.
  • tags — array — Tag slugs; match any by default.
  • limit — integer — Results per page, 1-100 (default 25).
  • match — string — How multiple `tags` combine: "any" (default, at least one) or "all" (every tag).
  • fields — array — Return only these keys on each record, e.g. ["slug","title","url"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • category — string — Vocabulary category: resources, actions, personas, domains, schemas or tags.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_feeds read-only

  The static machine-readable feeds every network site publishes (JSON, APIs.json, zero-auth) — the no-key alternative to this API.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_tags read-only

  Browse and rank the network tag taxonomy — every subject tag across 16k+ docs, by frequency or breadth (how many content types use it). Search with q. The map of what API Evangelist covers.

  Parameters:

  • q — string — Free text matched against tag labels, e.g. "governance".
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • sort — string — "count" (default, most-used first) or "breadth" (spread across the most content types).
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_tag read-only

  One tag with its per-type counts and everything it labels across the whole network (stories, areas, guidance, papers, …). Filter with types.

  Parameters:

  • slug (required) — string — Tag slug, as find_tags returns it, e.g. "apis".
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • limit — integer — Results per page, 1-100 (default 25).
  • types — array — Only return tagged items of these types, e.g. ["posts"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_related read-only

  Given any doc (type + slug), the most related content across the whole network, ranked by shared-tag relevance (rare, specific tags weigh most). "More like this," across stories, areas, guidance, papers, conversations, and more.

  Parameters:

  • type (required) — string — Content type of the item `slug` names — the plural path name: posts, services, solutions, tools, areas, papers, videos, conversations, guidance…, e.g. "posts".
  • slug (required) — string — Slug of the item to find related content for, within `type`, e.g. "2026-09-29-you-cannot-see-your-own-api-from-the-outside".
  • limit — integer — Results per page, 1-100 (default 25).
  • types — array — Only return related items of these types, e.g. ["papers","posts"].
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_stats read-only

  PRO — catalog intelligence: per-type counts, stories-per-year, and the top tags by reach across the network.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_company_insights read-only

  Browse profiled companies (Fortune 1000 + API providers) by name; ranked by technology-readiness signal. Free.

  Parameters:

  • q — string — Free text matched against company name, e.g. "chase".
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• insights_dimensions read-only

  Cross-section ranking of the 40 tracked investment dimensions (AI, cloud, governance, …) by company signal. Free.

  Parameters:

  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• insights_adoption read-only

  Services, tools, or standards ranked by how many companies adopt them. Free.

  Parameters:

  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • type — string — What to rank by adoption across the companies: "service" (default), "tool" or "standard".
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• get_company_insight read-only

  Full demand-side profile for one company: 40-dimension readiness, adopted stack, per-quarter history. Pro.

  Parameters:

  • slug (required) — string — Company slug, as find_company_insights returns it, e.g. "jp-morgan-chase".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• company_gaps read-only

  A company's weakest investment dimensions — where it is thin. Pro.

  Parameters:

  • slug (required) — string — Company slug, as find_company_insights returns it, e.g. "jp-morgan-chase".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• company_dimensions read-only

  PRO — a company's investment dimensions, each read AGAINST the measured cohort: the cohort median, mean and max, this company's rank and percentile, and an index where 1.0 is exactly typical. A raw dimension score is unreadable on its own; these are the columns that make it mean something.

  Parameters:

  • slug (required) — string — Company slug, as find_company_insights returns it, e.g. "jp-morgan-chase".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• company_peers read-only

  PRO — the companies most comparable to this one (same sub-sector where the data supports it, same industry otherwise), each with signal totals and top dimensions. The response says which basis it used.

  Parameters:

  • slug (required) — string — Company slug, as find_company_insights returns it, e.g. "jp-morgan-chase".
  • limit — integer — How many peers to return, 1-100.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• company_whitespace read-only

  PRO — what this company's peers run that it does not, ranked by how many peers carry it. The opening list for a vendor selling into the account.

  Parameters:

  • slug (required) — string — Company slug, as find_company_insights returns it, e.g. "jp-morgan-chase".
  • limit — integer — How many whitespace rows to return, 1-200.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• company_signals read-only

  PRO — the company's adopted stack entity by entity (services, tools, standards), each with how many companies across the cohort carry it and its radar ring.

  Parameters:

  • slug (required) — string — Company slug, as find_company_insights returns it, e.g. "jp-morgan-chase".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_entities read-only

  Browse the demand-side entity catalog — every service, tool and standard companies are measured against, ranked by how many carry it. Free. This is the left-hand side of the inversion: pick an entity here, then call entity_demand to see WHO.

  Parameters:

  • q — string — Substring match on the entity name.
  • kind — string — Entity kind: "service", "tool" or "standard".
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • min_companies — integer — Only entities carried by at least this many companies.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• entity_demand read-only

  PRO — THE INVERSION. Every other insights tool runs company -> stack; this one runs stack -> companies: which companies show demand for a given service, tool or standard, broken down by industry. This is the prospecting query — "who should I be selling this to" — and it exists in no report, bundle or page. Returns BOTH the authoritative company count and the companies it can actually name, with the basis difference explained.

  Parameters:

  • kind (required) — string — Entity kind: "service", "tool" or "standard".
  • slug (required) — string — Entity slug or display name — either resolves ("spark" and "Apache Spark" both work).
  • limit — integer — How many named companies to return, 1-500.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • industry — string — Restrict the named companies to one industry.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• find_measured_standards read-only

  The standards bodies API Evangelist has actually MEASURED — repositories, maturity, governance, coalition. Distinct from find_standards, which searches 613 catalog entries describing what a standard IS. Free.

  Parameters:

  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• standard_adoption read-only

  How far a standards body actually got: repositories published, how many graduated past unclassified and the resulting share, release cadence, contributor count — with the provenance of every figure and an explicit list of what is NOT measured. Free, because a specification is a public good.

  Parameters:

  • slug (required) — string — Standards body slug, as find_measured_standards returns it, e.g. "asyncapi".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• standard_coalition read-only

  Who is actually behind a standard: the governing bodies and their membership, the working groups, the member companies, and the named people. Carries the caveats where our machine-readable source disagrees with the body's own published roster. Free.

  Parameters:

  • slug (required) — string — Standards body slug, as find_measured_standards returns it, e.g. "asyncapi".
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• standard_repositories read-only

  Every repository a standards body publishes, with maturity, stars, forks, open issues and archived status. Free.

  Parameters:

  • slug (required) — string — Standards body slug, as find_measured_standards returns it, e.g. "asyncapi".
  • page — integer — Page number, 1-based (default 1). Use with `limit` to walk a long list.
  • limit — integer — Results per page, 1-100 (default 25).
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• validate_api read-only

  Lint an OpenAPI, AsyncAPI, Arazzo, or JSON Schema document against the curated best-of-breed ruleset (Spectral) and return the findings. Pass your own `ruleset` to run rules you own instead of the catalog — which is what you should do before gating on anything.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• api_coverage read-only

  Measure how much of an API description your governance actually inspects — coverage by section, each rule's reach, and the rules that match nothing. Like test coverage, one altitude up.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• reconcile_waivers read-only

  Reconcile a waivers file against lint output: suppress sanctioned violations and surface the expired, stale, and expiring ones. Returns the effective failing set a build should gate on.

  Parameters:

  • waivers (required) — ['string', 'object'] — The waivers file to reconcile — sanctioned violations, with expiry.
  • violations (required) — ['array', 'object', 'string'] — Lint output to reconcile against the waivers. An array of findings, or the object a linter returned.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• governance_report read-only

  Turn Spectral findings into a self-contained HTML governance report — severity summary, grouped by rule, framed toward progress.

  Parameters:

  • results (required) — ['array', 'object'] — Spectral findings to render. An array, or the `{ results: [...] }` object a run returned.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• verify_certificate read-only

  Re-verify a governance certificate against the API it attests to. If the API drifted or the certificate was forged, verification fails.

  Parameters:

  • certificate (required) — ['string', 'object'] — The governance certificate to re-verify. Alias: `cert`.
  • api (required) — ['string', 'object'] — The API description being attested or verified. String or parsed object.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• classify_api_fields read-only

  Inventory the PII, secret, and sensitive fields an OpenAPI exposes, across parameters and request/response schemas. Run this before exposing an API to agents.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• diff_api_versions read-only

  Compare two OpenAPI versions and classify every change as breaking, non-breaking, or an addition — at path, operation, parameter, request-body, and response granularity.

  Parameters:

  • old (required) — ['string', 'object'] — The BASELINE version of the document. Aliases: `from`, `base`.
  • new (required) — ['string', 'object'] — The CANDIDATE version to compare against the baseline. Aliases: `to`, `revision`.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• api_scorecard read-only

  Score an API's maturity 0-100 with a letter grade across design, governance coverage, documentation, and (from an apis.json) discoverability, agent-readiness, and operations. Pass your own `ruleset` — a score against rules you never adopted means little.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• govern_estate read-only

  PRO — Run the core governance checks across a WHOLE SET of internal APIs and roll them up: per-API score/grade/errors/PII, plus estate aggregates (avg score, grade distribution, APIs with PII/secrets) and a worst-first punch-list of what to fix. Pass `apis` ([{id, document}]) and your own `ruleset`. detail:true adds per-API dimensions.

  Parameters:

  • apis (required) — array — The estate: one entry per API, [{ id, document }]. `document` alone is accepted for a single API.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• deprecation_posture read-only

  Report every deprecated operation and whether it actually announces a sunset. The ones that go quiet are the ones that break consumers.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• validate_onboarding read-only

  Validate an API Onboarding Descriptor (AID) against the api-commons schema and report which optional sections it fills in.

  Parameters:

  • descriptor (required) — ['string', 'object'] — An API Onboarding Descriptor (AID) to validate. Aliases: `aid`, `document`.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• certify_api read-only

  Issue a tamper-evident governance certificate attesting an API passed a named ruleset at a profile threshold. A SHA-256 fingerprint lets anyone re-verify it.

  Parameters:

  • api (required) — ['string', 'object'] — The API description being attested or verified. String or parsed object.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• export_agent_rules read-only

  Turn a governance ruleset into agent-native guidance — an AGENTS.md block, a system-prompt set, a per-rule remediation pack, and a digest — so agents follow the rules while authoring, not after linting.

  Parameters:

  • ruleset (required) — object — A Spectral ruleset to run instead of the catalog default. Score against rules you never adopted means little — pass your own.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• audit_pipeline read-only

  Score a repository's Spectral CI setup against an 8-point maturity rubric and return a prioritized punch-list. Pass a public `repo` ("owner/name") or `files`.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world yes

• forge_mcp_tools read-only

  From an OpenAPI, forge an MCP server design, per-operation tools, Agent Skills, an enriched OpenAPI, a server.json, and apis.json properties.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• gate_agent_context read-only

  Choose which operations and fields an API exposes to agents and emit a governed Tyk OpenAPI, an MCP tool manifest, and a Spectral ruleset with PII and secret checks.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• apply_overlay read-only

  Apply an OpenAPI Overlay to a document and report what each action matched. Use it for deprecation and migration choreography without forking the contract.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • overlay (required) — ['string', 'object'] — An OpenAPI Overlay document whose actions are applied to `document`.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• mock_payloads read-only

  Generate an example request and success-response payload for every operation from its schema.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• code_snippets read-only

  Generate runnable curl, JavaScript, and Python samples for every operation, with path params filled in and a realistic body.

  Parameters:

  • document (required) — ['string', 'object'] — The API description to operate on — OpenAPI, AsyncAPI, Arazzo or JSON Schema. A YAML/JSON string or an already-parsed object; both are accepted.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• score_reusability read-only

  Score how reusable a set of APIs is and detect duplication across them — the anti-sprawl view of an estate.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• map_api_experience read-only

  Map each REST operation to its MCP tool and Agent Skill — the API -> MCP -> Agent-Skill chain — from an APIs.json, with a free/paid coverage scorecard and a flow map.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• generate_agent_descriptor read-only

  Generate an API's llms.txt and an AGENTS.md consumption contract — base URL, auth, the operation table, deprecated operations to avoid, and usage rules.

  Parameters:

  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.

  Hints: read-only yes · destructive not declared · idempotent not declared · open-world no

• submit_api_evangelist_feedback

  Report something wrong with a result: wrong data, a filter that did not filter, a ranking that put off-topic results first, a number you could not reproduce, or an answer that was useless for your task. Free, and the fastest way to get a defect fixed — the support sweep reads the log this lands in. Say what you expected as well as what you got.

  Parameters:

  • problem (required) — string — What was wrong. Be specific — the tool you called, the arguments, and what came back.
  • tool — string — Optional name of the tool that produced it.
  • about — string — Optional slug, area, paper or story the report concerns.
  • context — string — Optional: why you are asking. One sentence — the task you are trying to complete, or what you expect to get back. Never included in the answer and never used to rank; it is read only when a result turns out to be wrong, which is when knowing the intent is what makes the report actionable.
  • expected — string — What you expected instead.
  • severity — string — wrong = factually incorrect; misleading = correct but reads as something else; unhelpful = right and useless.

  Hints: read-only no · destructive no · idempotent no · open-world no

Publisher website Repository

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.