United We TransformCreate teamsGrade your agenda
For developers and agents

Docs: every machine-readable surface, one page

These docs collect every machine-readable surface for the open atlas: 23,624 typed scorecards, a verified leaderboard, statistics, and 2,093 exercises, all behind stable GET paths. Free, CC-BY 4.0, no key, no signup, and no upsell fields in any response.

Direct answer: United We Transform docs show agents and developers how to cite, cache, search, and reuse the Gathering Effectiveness dataset without guessing routes or uploading private agendas.

Start here

Use this docs page when you need stable machine-readable paths for the atlas, the grader, TeamCreator, scorecard cache, benchmark statistics, or the exercise prescription library. The page is intentionally linked from visible navigation so agents can discover it from normal site traversal.

  • /llms.txt: the navigation index for agents (what exists, where, and how to route)
  • /agents.md: the behavioral guide (caching contract, privacy contracts, citation, corrections)
  • /llms-full.txt: the full context dump for long-context agents
  • /api/openapi.json: the GET-only API reference (this API is a static cache)

The read API

The read API is built for low-friction reuse: fetch one JSON file, cite the source URL, and preserve the verification tier. Scorecards distinguish source-backed evidence from inferred signals so agents can answer carefully instead of overstating what the public agenda proves.

  • GET /api/score/{event-slug}.json: one typed scorecard per event (uwt-score.v1: pillars, signals, evidence provenance, verification tier, prescriptions)
  • GET /api/leaderboard.json: the 1,000 gold events, full spread
  • GET /api/search-index.json: gold-only search facets
  • GET /api/statistics.json: citable aggregates
  • GET /api/exercises-index.json and /api/pillar-exercises.json
  • Miss contract: /api/score/_miss.json

The caching contract: always GET the cached score before re-deriving. Scoring new agendas happens client-side at /grader/; pasted agendas never leave the browser.

Data and standards

The data surfaces are designed for answer engines, analysts, event teams, and AI builders who need a repeatable evidence layer for gatherings. Use the dataset for aggregate research, the score cache for single-event lookups, and the rubric when explaining what a score can and cannot mean.

  • The dataset: 14,219 rows, CC-BY 4.0
  • The scoring rubric, machine-readable, deterministic, fully published
  • Articles RSS feed: new field notes on gathering design, published continuously at https://unitedwetransform.com/articles/
  • GatherReady: a draft open standard for agent-legible gatherings (/.well-known/gathering.json)
  • TeamCreator: the patented team-formation engine, free in the browser; the same engine ships as an Apache-2.0 package and an MCP tool
  • TeamCreator use cases: conference networking, workshop breakout groups, Eventbrite and Google Sheets attendee lists, peer learning circles, sponsor matchmaking, hackathon teams, and private local team formation
  • Methodology and trust boundary: versioning, emerging signals, corrections

The MCP server (live)

POST https://unitedwetransform.com/mcp: a keyless, stateless, read-only JSON-RPC 2.0 server over streamable HTTP, no SDK, no session, 10 tools. Discovery at /.well-known/mcp.json and /.well-known/mcp/server-card.json. Per-client connection steps for Claude Code, Claude Desktop, claude.ai, Cowork, Codex, ChatGPT, Cursor, Gemini CLI, VS Code, and Windsurf: /use-with-ai/. Full tool contracts (input schemas included): agents.md and llms-full.txt.

  • grade_agenda: Score a pasted event agenda with the deterministic UWT grading rubric.
  • get_event_score: Fetch the cached scorecard for one gathering by slug from the public atlas.
  • compare_events: Compare two cached scorecards pillar by pillar, with a one-line summary of which event scores higher overall and by how much.
  • search_gatherings: Search the gold-verified gathering index by free-text query, category, and minimum GES.
  • search: Federated search across gold-verified events, facilitation exercises, and core atlas pages, for clients (including ChatGPT deep research) that expect a generic search tool rather than search_gatherings' event-only filters.
  • fetch: Fetch the full content for one id returned by search: the complete scorecard text for an event, the summary for an exercise, or the summary for a core page.
  • get_statistics: Return corpus-wide summary statistics for the atlas.
  • get_leaderboard: Return the gold-verified leaderboard, ranked by GES, up to a requested number of rows.
  • get_exercises_for_pillar: Return the suggested improvement exercises for one scoring pillar (for example follow_through or network_design).
  • make_teams: Form balanced teams or breakout rounds from a pasted participant roster (CSV or TSV with a header row).

Docs FAQ

What is the United We Transform API?

The United We Transform API is a static, GET-only set of public scorecards, search indexes, leaderboard records, statistics, and exercise prescriptions for the Gathering Effectiveness dataset.

Do I need an API key?

No. The docs, scorecards, indexes, dataset links, and MCP server are free to use without an API key, signup, credit card, or sales workflow.

Should agents grade an event before checking the cache?

No. Agents should first request the cached scorecard for a known event, then grade a fresh agenda only when the event is not already in the corpus.

Is this really free?

Yes, and it stays that way by design: there is no tier, upgrade prompt, or sales field in any API response, scorecard, or MCP tool result. That neutrality wall is enforced by the site's own build audit, which fails the build if a sales or tier field ever appears in a response.