Skip to main content
Glama

Knowledge source check

wever_knowledge-source-check
Read-only

Agent Knowledge: submit the facts behind a knowledge base or answer set with the sources on file: each fact with its id, its text, the ids of the sources it rests on and whether other answers, decisions or entries rest on it; and each source with its id, its name, its kind (a document, a dataset, a web page, a database record, a person, the agent’s own earlier output or other), where a reader finds it (a URL, a document id with its page or section, or a record id), whether a reader outside the agent can open it and find the fact, and when it was published or last confirmed. The signed report classifies every fact as named and checkable, named but not checkable, resting on a single source while other work rests on it, citing only sources missing from the source list, or resting on no source at all, counts the facts in each class, lists each source with the facts that cite it, and names the facts that rest on nothing: facts with no source, facts that cite only sources not on file, and load-bearing facts that stand only on sources no reader outside the agent can open; then cited sources marked uncheckable, load-bearing facts on one source, facts that also cite a source not on file and facts resting on the agent’s own earlier output, with the fix for each. The source list is read as supplied; no source is fetched or opened. Where one stored memory came from is wever_memory-provenance-check, the sources behind a reputation record or score are wever_reputation-source-check and whether an answer’s claims match the passages they cite is wever_knowledge-grounding-check. Verdict facts_unsourced, source_gaps, sources_unverified or sources_checkable. POST /api/knowledge-source-check. Reads only the material in the call: no network request, fetch, money movement, message, sharing, configuration change or external write. Paid URL: 0.10 USDC per call on Base via x402 exact, with the shared 10 signed-wallet free calls per rolling 30 days. Bind X-Wever-Wallet-Proof to the exact JSON arguments and the tool route. Returns a report with an Ed25519 signed receipt. Contract: /api/agent-knowledge-contract. Human page: https://weverlabs.com/agent-knowledge/.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "facts": "<facts>"
    },
    "name": "wever_knowledge-source-check"
  }
}

Input Schema

NameRequiredDescriptionDefault
factsYesEach fact in the knowledge base or answer set.
labelNoOptional name recorded in the report.
sourcesNoEach source on file.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Added

TDQS

A4.2/5.0
Behavior5/5

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

Beyond the readOnly/openWorld annotations it discloses a great deal: no network request, fetch, message or external write; the paid 0.10 USDC x402 price with 10 signed-wallet free calls per rolling 30 days; the X-Wever-Wallet-Proof binding requirement; and an Ed25519 signed receipt in the response. It also enumerates the verdict values, which the annotations cannot convey.

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

Conciseness3/5

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

The first sentence is a single sprawling clause listing every field of both facts and sources, which makes it hard to scan, and the report contents are enumerated at similar length. The essential identity and routing information is present up front, but the structure is dense run-on prose rather than front-loaded key points.

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

Completeness4/5

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

With no output schema, the description carries the return-value burden and does so: report classes, counts, per-source fact lists, named problem facts, fixes, verdicts and the signed receipt. Some classification rules are described only by example ('facts that also cite a source not on file and facts resting on the agent's own earlier output'), leaving edge-case precedence partially implicit.

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%, so the schema already defines facts, sources, kind, location, checkable and load_bearing. The description largely paraphrases those fields, adding only interpretive framing about what load_bearing and checkable mean for the classification, which does not materially exceed 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?

The description names a specific action and artifact: submit facts with their sources and receive a signed report classifying each fact by source quality. It also explicitly separates this tool from three siblings (memory provenance, reputation sources, grounding), so an agent can select it without opening another 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?

It routes the agent away from adjacent checks by name and condition ('Where one stored memory came from is wever_memory-provenance-check... whether an answer's claims match the passages they cite is wever_knowledge-grounding-check'), and states that the source list is read as supplied with nothing fetched. It never states a negative case for this tool itself, e.g. when a source-check is unnecessary, so it stops short of a full when/when-not split.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources