Likelier MCP

Citation-grade probability data — ~457 real-world risks + ~179 decision regret-pairs across 44 locales — exposed as an MCP server. Use from Claude Desktop, Claude Code, Cursor, Continue, or any MCP-compatible client.

The MCP wraps the same dataset that powers /risks and /decisions on this site — every claim comes with a verbatim source excerpt, uncertainty bounds, and a Wayback archive URL. The tool calibrate_risk takes a free-text situation and returns grounded retrieval (matched risks + scope-matched anchors + applicable demographic multipliers).

Install (remote — recommended)

One config line. No install, always-current data. Add to your client's MCP config:

Claude Desktop / Claude Code

{
  "mcpServers": {
    "likelier": {
      "url": "https://likelier-mcp-remote-staging.krzysztof-gluszczyk.workers.dev/mcp"
    }
  }
}

Claude Desktop config: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows). Claude Code: project .mcp.json.

The hostname above is the current staging URL — production domain (mcp.likelier.com) coming with launch.

Cursor / Continue

Same URL; consult each tool's MCP configuration docs for the exact JSON field name (typically url under mcpServers).

Advanced: local / offline use

v1 ships remote MCP only. The remote URL above works with every MCP client we've tested (Claude Desktop, Claude Code, Cursor, Continue), data refreshes every 5 minutes against the R2 snapshot, and every response carries the citation envelope + (where applicable) a feedback URL.

If you need a local/offline pass-through — for hacking on the server, working without internet, or pinning a specific snapshot — clone the source from GitHub and run the stdio entrypoint directly with LIKELIER_SNAPSHOT_URL=file:///path/to/snapshot.json. Source: https://github.com/kgluszczyk/likelier-mcp (repo public with v1 launch).

Caveat: a locally-cached snapshot only refreshes on process restart, so it can drift from the published corrections log. Prefer the remote endpoint for any production workflow.

Tool surface (v1)

calibrate_risk(situation, locale?, user_context?) — primary decision-support entry point. Takes a free-text situation ("I ate moldy bread", "should I get LASIK"), returns matched risks + scope-matched anchors + applicable multipliers. Hard-refuses medical / legal / financial categories with a "consult a licensed professional" message.
get_risk(slug, locale?) — full entry: probability, sources, uncertainty, multipliers
list_risks({ category?, limit? }) — filtered list (default limit 5)
search_risks(query, limit?) — BM25 lexical search
risks_by_probability({ min?, max?, ... }) — range queries + scope/uncertainty filters
find_anchor_risks(probability, k?) — "what's roughly as likely as X"
similar_risks(slug, mode?) — modes: same_probability, semantic, different_category
compare_risks(a, b) — side-by-side with scope-warning
get_citation(slug, source_index?, style?) — verbatim source excerpts
Decisions mirror: get_decision, list_decisions, search_decisions, decisions_by_regret_rate, similar_decisions, compare_decisions
suggest_data_gap(question, kind, situation?, missing_field?, contact?) — file a structured suggestion when the dataset lacks coverage. On hosted MCP these land in the issue tracker automatically; on local stdio the payload is printed to stderr.
Resource: methodology://about

Trust stack

Every tool response includes a dataset_version envelope. Source excerpts are wrapped in <verbatim_source_text> blocks so the calling LLM treats them as data rather than instructions (prompt-injection defense). Scope mixing (e.g. activity-specific vs lifetime probabilities) is flagged with scope_warning: true on sorted/filtered lists.

For non-MCP consumers

The raw JSON the MCP wraps is also published statically — useful for bulk consumers, researchers, or non-MCP integrations:

/api/fears.json — list of all reviewed risks
/api/fears/<slug>.json — single entry
/api/decisions.json — decision regret-pairs
/api/categories.json — taxonomy
/llms.txt — LLM-facing site description

Found a gap? Tell us

The dataset has ~457 risks + ~179 decisions today — nowhere near every real-world risk a person might ask about. If calibrate_risk comes back with no_match, or an entry is missing a multiplier / scope / better source you have on hand, surface it. Two paths:

In-flow — ask your agent to call the suggest_data_gap(question, kind, situation?, missing_field?, contact?) tool. On the hosted MCP, suggestions land in the issue tracker automatically.
Direct — open the "Suggest a missing risk or decision" GitHub issue template. Better fit when you already have sources lined up.

Good suggestions cite at least one authoritative source (peer-reviewed, government report, primary data, reputable reference). News articles alone aren't enough to build an entry on. Probability or scope errors with a superseding citation are equally welcome.

License

MCP package code: MIT. Dataset: CC-BY-SA 4.0 — attribution to Likelier.com.

Support

Likelier is free and community-funded. If the MCP saved you time or shipped a feature for you, a coffee buys hosting + content curation:

If this helped you think more clearly about risk, consider supporting the project so we can keep the data trustworthy and ad-free.

Agent-native risk data