Skip to main content
Glama

corroborate_sources

Read-onlyIdempotent

Checks whether retrieved web or RAG snippets independently support a factual claim by collapsing syndicated duplicates and flagging cross-domain contradictions before you answer.

Instructions

Evaluate independent evidentiary support across retrieved RAG or web snippets by collapsing syndicated wire-copy duplicates and checking cross-domain contradictions. Call before answering factual queries from retrieved passages; do not use to fetch web URLs (URLs are parsed offline solely for domain grouping and never contacted over the network) or to audit tool-execution prose (use verify_narration). Supply at least two sources from distinct domains to reach multi-source proof standards (sources=[] yields speculation with meets_casual_burden=false); contradictions set disputed=true without failing, whereas an empty query or malformed sources item returns isError=true.

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": {
      "query": "<query>",
      "sources": "<sources>"
    },
    "name": "corroborate_sources"
  }
}

Input Schema

NameRequiredDescriptionDefault
queryYesUser question or claim being answered from retrieved sources.
sourcesYesRetrieved web or RAG snippets with source URLs.

Output Schema

NameRequiredDescriptionDefault
hazardYesClassified risk tier of query scaling required corroboration.
disputedYesTrue if cross-domain NLI detected conflicting claims.
dispute_sourcesYesDomain or URL pairs exhibiting NLI contradiction.
achieved_standardYesHighest evidentiary standard met after deduplication.
raw_sources_countYesTotal input snippet items supplied in sources.
meets_casual_burdenYesTrue if achieved_standard satisfies required_casual_standard.
required_casual_standardYesMinimum proof standard required for hazard tier.
independent_sources_countYesRemaining independent source clusters after deduplication.
deduplicated_duplicates_collapsedYesCount of mirrored or syndicated passages collapsed.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv0.1.2
    • changedInput schema / properties / sources / items / properties / url / description
      Previous value: -"Canonical URL or domain identifier of the passage."New value: +"Source URL or hostname parsed offline for domain."
    • changedOutput schema / (root)
      Previous value: -nullNew value: +{
      +  "additionalProperties": false,
      +  "properties": {
      +    "achieved_standard": {
      +      "description": "Highest evidentiary standard met after deduplication.",
      +      "enum": [
      +        "speculation",
      +        "balance of probabilities",
      +        "clear and convincing",
      +        "beyond reasonable doubt"
      +      ],
      +      "type": "string"
      +    },
      +    "deduplicated_duplicates_collapsed": {
      +      "description": "Count of mirrored or syndicated passages collapsed.",
      +      "type": "integer"
      +    },
      +    "dispute_sources": {
      +      "description": "Domain or URL pairs exhibiting NLI contradiction.",
      +      "items": {
      +        "type": "string"
      +      },
      +      "type": "array"
      +    },
      +    "disputed": {
      +      "description": "True if cross-domain NLI detected conflicting claims.",
      +      "type": "boolean"
      +    },
      +    "hazard": {
      +      "description": "Classified risk tier of query scaling required corroboration.",
      +      "enum": [
      +        "low",
      +        "medium",
      +        "high"
      +      ],
      +      "type": "string"
      +    },
      +    "independent_sources_count": {
      +      "description": "Remaining independent source clusters after deduplication.",
      +      "type": "integer"
      +    },
      +    "meets_casual_burden": {
      +      "description": "True if achieved_standard satisfies required_casual_standard.",
      +      "type": "boolean"
      +    },
      +    "raw_sources_count": {
      +      "description": "Total input snippet items supplied in sources.",
      +      "type": "integer"
      +    },
      +    "required_casual_standard": {
      +      "description": "Minimum proof standard required for hazard tier.",
      +      "enum": [
      +        "speculation",
      +        "balance of probabilities",
      +        "clear and convincing",
      +        "beyond reasonable doubt"
      +      ],
      +      "type": "string"
      +    }
      +  },
      +  "required": [
      +    "hazard",
      +    "achieved_standard",
      +    "raw_sources_count",
      +    "independent_sources_count",
      +    "deduplicated_duplicates_collapsed",
      +    "disputed",
      +    "dispute_sources",
      +    "required_casual_standard",
      +    "meets_casual_burden"
      +  ],
      +  "type": "object"
      +}
  2. First observedv0.1.0

TDQS

A4.9/5.0
Behavior5/5

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

Beyond the annotations, the description discloses that URLs are parsed offline solely for domain grouping and are never contacted over the network, which is important context for an openWorldHint=false tool. It also describes non-obvious outcomes: sources=[] yields speculation with meets_casual_burden=false, contradictions set disputed=true without failing, and empty/malformed input returns isError=true. These details materially help an agent predict behavior.

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

Conciseness5/5

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

The description is dense but front-loaded, beginning with the core action and then moving through usage, exclusions, source requirements, and outcome semantics. Every sentence carries operational information, and there is no filler or repetition.

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?

Given that the schema is fully described, an output schema exists, and annotations already declare the safety profile, the description is complete enough for correct invocation. It covers source requirements, duplicate handling, contradictions, error behavior, and offline URL treatment, leaving no major decision-relevant gap for this tool.

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

Parameters4/5

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

Schema coverage is 100%, so the baseline is 3; however, the description adds meaning not obvious from the schema. It explains that at least two sources from distinct domains are required for multi-source proof standards, that sources=[] has a specific speculative outcome, and that URLs are used only for offline domain grouping. This adds useful semantic constraints around the parameters.

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 uses a specific verb and resource: evaluate independent evidentiary support across retrieved RAG or web snippets, including duplicate collapsing and cross-domain contradiction checks. It clearly distinguishes the tool from verify_narration and from URL-fetching behavior, so an agent can identify its role without opening sibling schemas.

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

Usage Guidelines5/5

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

The description explicitly states when to call it: before answering factual queries from retrieved passages. It also states exclusions: do not use it to fetch web URLs and do not use it to audit tool-execution prose, directing the agent to verify_narration instead. It further specifies the source requirements and outcomes for insufficient sources, contradictions, and malformed input.

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