Trent

Get context provenance

get_context_provenance

Read-only

Show which Trent results reference the customer's own material.

For one scan, group findings, recommendations, and security requirements
by the exact version of the document or tool report they reference.
Sources can be uploaded, pushed by CI, or synced by a connector. Use this
to answer "where did the file I uploaded get used?" or
"does this version still matter?".

This tool gives a read-only view of recorded references. Describe them
as "Trent referenced this passage when producing this finding". Do not claim the
document caused a finding or that Trent would have missed it without the
document; Trent has not run that comparison. When provenance_recorded is
false, say the scan kept no record of what it read. A missing record does
not prove that the scan used no context.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "get_context_provenance"
  }
}

Input Schema

Name	Required	Description	Default
`cursor`	No	Opaque cursor from a previous call's `next_cursor`.
`moment`	No	Which scan to answer from: 'current' (the newest completed one), 'settled' (the same, but refused while a scan is publishing), or 'scan:{scan_id}' to pin one scan.	current
`project_id`	No	Project ID (provide either project_id or project_name)
`revision_id`	No	Narrow the answer to one exact source version, as returned in a previous call or on a result's provenance. Omit for every version this scan read.
`project_name`	No	Project name (provide either project_id or project_name)

Output Schema

Name	Required	Description
`error`	No	True when the call itself failed. False for successful calls, including a scan still running, an already approved plan, or a repository classified as non-agentic. The reason field describes these workflow states.
`reason`	No	A stable snake_case code from the allowed set. Present on every error and on successful calls that carry a workflow state.
`message`	No	Exactly one sentence written for the user. Never raw exception text or a backend error string, which may expose private details.
`scan_id`	No	The scan the answer is taken from.
`metadata`	No	Caller metadata, such as 'expiration_warning'. Substantive results have their own fields. Never includes exception or backend error text, which may expose private details.
`revisions`	No	One entry per version of the customer's material this scan accounts for.
`next_steps`	No	Recommended next actions.
`project_id`	No	Resolved Trent project identifier.
`next_cursor`	No	Pass back as `cursor` to read the next page. Absent when the last page was served.
`provenance_recorded`	No	False when no record can be shown. reason says why: no_analysis, the scan kept no record (it predates provenance recording or read none); no_completed_scan; scan_running; or scan_not_in_project. Report the missing record. It does not establish that the scan used no customer context.
`retry_after_seconds`	No	With reason rate_limited, how many seconds Trent asks the caller to wait.

Schema Changelog

Changes observed during successful MCP inspections.

Changed6 schema fields changedOct 30, 2026

changedOutput schema / $defs / ProvenanceRevision / properties / claim_counts / description

Previous value: -"For external findings: how many of the file's findings Trent confirmed or dismissed."New value: +"For external findings: how many of the file's findings Trent confirmed or dismissed. Dismissed findings are the ones get_security_posture lists as Rejected, with Trent's reason, in external_findings.unadopted_claims."

addedOutput schema / $defs / ProvenanceRevision / properties / limit_bytes

Added value: +{
+  "anyOf": [
+    {
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "With failure_reason size_limit, the limit it exceeded, when Trent recorded it.",
+  "title": "Limit Bytes"
+}

addedOutput schema / $defs / ProvenanceRevision / properties / size_bytes

Added value: +{
+  "anyOf": [
+    {
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "With failure_reason size_limit, the source's size, when Trent recorded it.",
+  "title": "Size Bytes"
+}

changedOutput schema / $defs / ProvenanceRevision / properties / status / description

Previous value: -"ready, failed, anchors_unavailable, or carried. carried means an earlier scan read the source. Attribute that read to the earlier scan."New value: +"ready, failed, anchors_unavailable, or carried. carried means an earlier scan read the source. Attribute that read to the earlier scan. anchors_unavailable means Trent read the source but has no passages in it to cite; for an external findings file this is normal, since its findings reach results as claims (claim_counts), not as cited passages."

changedOutput schema / properties / provenance_recorded / description

Previous value: -"False when the scan kept no record of the customer material it read, either because it predates provenance recording or read none. Report the missing record. It does not establish that the scan used no customer context."New value: +"False when no record can be shown. reason says why: no_analysis, the scan kept no record (it predates provenance recording or read none); no_completed_scan; scan_running; or scan_not_in_project. Report the missing record. It does not establish that the scan used no customer context."

addedOutput schema / properties / retry_after_seconds

Added value: +{
+  "anyOf": [
+    {
+      "type": "integer"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "With reason rate_limited, how many seconds Trent asks the caller to wait.",
+  "title": "Retry After Seconds"
+}

First observedOct 23, 2026

TDQS

A4.1/5.0

Behavior5/5

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

Beyond the readOnlyHint/openWorldHint annotations, it discloses that this is a view of *recorded* references only, that causation must not be inferred ('Trent has not run that comparison'), and how to read provenance_recorded=false. That is exactly the behavioral context an agent needs to avoid overclaiming from the output.

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 interpretive caveats. Slightly long, but the cautionary sentences about causation and missing records earn their place given how easily the output could be misread.

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?

Output schema exists, so return values needn't be described, and annotations plus full schema coverage handle safety and parameters. Combined with the scoping and interpretation guidance, an agent has everything needed to call and correctly frame the result.

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 cursor, moment, revision_id, and the project identifiers are already documented. The description reinforces 'for one scan' and 'exact version' conceptually but adds no syntax or semantics beyond the schema, so baseline 3 applies.

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

Purpose4/5

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

States a specific verb and resource: groups findings, recommendations, and security requirements by the exact source version they reference for one scan. The scope ('which Trent results reference the customer's own material') is distinct from retrieval tools like get_artifact, but the description never names the closest sibling (find_related_sources), so a direct contrast is left to inference.

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 explicit motivating questions — 'where did the file I uploaded get used?' and 'does this version still matter?' — which make the when-to-use clear. It lacks an explicit when-not or an alternative tool to prefer, which keeps it shy of a 5.

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

All Trent tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.