facts-mcp

Overview Schema Source Related Servers Discussions Admin

Query the registry

facts.recall.query

Read-onlyIdempotent

Read recorded claims from a project registry with timestamps, staleness marks, and remeasure instructions; retracted claims stay hidden unless requested.

Instructions

Read claims from the project registry. It NEVER measures and opens no sockets: it returns what was recorded, each claim with its own age and stale mark, plus remeasure_with — the tool and arguments to measure it again yourself. Retracted claims are hidden unless asked for; they are never gone.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`tags`	No	Conjunction: all of them must be present. Normalized as on assert.
`text`	No	Case-insensitive substring over statement. No ranking, no relevance.
`limit`	No	Page size.
`order`	No	Sort order.	observed_desc
`cursor`	No	Opaque cursor from a previous page. Every page is cut from the registry as it stood at the first one, so claims written meanwhile never shift the window. Valid only with the filters and order it was issued for.
`project`	Yes	Project whose registry to read.
`evidence`	No	How much evidence to include. `ref` keeps responses small.	ref
`observed_by`	No	Exact match on the author.
`stale_seconds`	No	Threshold for the stale mark. Your threshold, not a judgement of ours. Defaults to the server's configured one (an hour unless set).
`observed_after`	No	ISO 8601 lower bound on observed_at.
`only_retracted`	No	Only retracted claims. Takes precedence over include_retracted.
`observed_before`	No	ISO 8601 upper bound on observed_at.
`include_retracted`	No	Bring retracted claims too.

Output Schema

TableJSON Schema

Name	Required	Description
`ok`	Yes	Whether the observation could be taken. NOT a judgement about what was observed: a cleanly received HTTP 500 is ok:true.
`data`	Yes	The raw observation, or null when error is set.
`tool`	Yes	Full tool name.
`error`	Yes	What prevented observing, if anything.
`target`	Yes	Effective arguments, redacted.
`project`	Yes	The normalized project.
`warnings`	Yes	Non-fatal notices.
`age_basis`	Yes	Which clock the age came from.
`request_id`	Yes	The request that produced it.
`age_seconds`	Yes	Seconds since observed_at.
`duration_ms`	Yes	How long the observation took.
`observed_at`	Yes	ISO 8601 with offset.
`observation_id`	Yes	Citable id of this observation.

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.3.0Oct 21, 2026

TDQS

A4.2/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint, idempotentHint and openWorldHint=false, so the safety profile is covered; the description nonetheless adds real value by confirming no socket activity and disclosing output traits — per-claim age and stale mark, a remeasure_with field, and retraction visibility semantics. It does not discuss permissions, limits beyond paging, or failure behavior, keeping it short of a 5.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the core action, then the non-measurement guarantee, then return and retraction behavior — a sensible order. Three sentences with little waste, though the read-only point is restated in two guises ("NEVER measures" / "returns what was recorded"), which slightly dilutes it.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

With 13 fully documented parameters and an output schema present, the description only needs to supply the behavioral frame, and it does: read-only, closed-world, snapshot-consistent paging, and non-destructive retraction. Nothing an agent needs to invoke this correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100% and each of the 13 parameters has a detailed inline description, so the schema does the heavy lifting and a 3 is the correct baseline. The prose only adds conceptual color about retracted claims, not syntax or semantics beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ("Read claims from the project registry") and immediately delimits scope against the measure siblings: "It NEVER measures and opens no sockets: it returns what was recorded." An agent can distinguish this from facts.measure.http/ws and the other recall tools without opening a schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear context for use — read recorded claims rather than produce fresh measurements — and covers a key behavioral branch ("Retracted claims are hidden unless asked for"). It never explicitly names an alternative tool or the condition that selects facts.recall.history/assert/retract, so routing is implied rather than spelled out.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server