Trent

Correct a feedback note

update_feedback

Destructive

Replace an existing feedback item with corrected text. To remove an item without replacing it, use remove_feedback; to add new feedback, use record_feedback.

The call removes the earlier item. Use a feedback_id from a prior
list_feedback result. Do not guess or reconstruct an id; it identifies
the item this call removes.

Report redaction_outcome and reach. 'already_redacted' means the item
was gone before the call. 'failed' means it remains after a removal
failure. Describe that outcome to the user.

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": {
      "feedback_id": "<feedback_id>",
      "content": "<content>"
    },
    "name": "update_feedback"
  }
}

Input Schema

Name	Required	Description
`content`	Yes	The correction the user expressed in their current message, in their own words. Carry over earlier text when the user asked to merge it.
`stage_id`	No	Which analysis stage the replacement belongs to. Omit to inherit the stage of the item being corrected.
`project_id`	No	Project ID (provide either project_id or project_name)
`feedback_id`	Yes	Identifier of the feedback item to replace, from a list_feedback result.
`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.
`reach`	No	Meaningful when redaction_outcome is 'redacted'. 'not_yet_read': the earlier version was unread, so the next analysis uses the correction. 'may_have_been_read': a scan may have read the earlier version and will finish with it; the correction applies from the next run. Neither value confirms retraction from an analysis already running.
`reason`	No	A stable snake_case code from the allowed set. Present on every error and on successful calls that carry a workflow state.
`item_id`	No	Lasting identifier of the item. It stays the same across corrections.
`message`	No	Exactly one sentence written for the user. Never raw exception text or a backend error string, which may expose private details.
`version`	No	Version of the item this call wrote, one above the earlier item's.
`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.
`next_steps`	No	Recommended next actions, in order.
`feedback_id`	No	Identifier of the new item that replaces the earlier one.
`redaction_outcome`	No	Outcome for the earlier item. 'redacted': this call removed it. 'already_redacted': it was removed before this call, so list the project's feedback to check for an existing replacement before recording more. 'failed': removal failed and the item remains; remove_feedback can remove it separately.
`retry_after_seconds`	No	With reason rate_limited, how many seconds Trent asks the caller to wait.
`replaced_feedback_id`	No	Identifier of the earlier item this correction replaces.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedOct 30, 2026

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

A4.3/5.0

Behavior4/5

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

Annotations already flag destructiveHint=true and idempotentHint=false, and the description is consistent with them, adding what actually happens ('the call removes the earlier item') plus failure semantics. It enumerates redaction_outcome values ('already_redacted', 'failed') and tells the agent to surface the outcome to the user — useful behavior not derivable from the annotations. Slight overlap with the output schema keeps it below 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?

Three short paragraphs, front-loaded with the verb and the sibling alternatives before the mechanics. The middle paragraph partially repeats the schema's feedback_id documentation, which is the only wasted space.

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 annotations covering the destructive profile and an output schema covering return shape, the description supplies the remaining decision-critical pieces: sibling routing, id provenance, and how to interpret failure outcomes. It omits any mention of stage_id/project scoping, though the schema covers those, so nothing is truly 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%, so content, stage_id, project_id/project_name and feedback_id are all documented in the schema itself, establishing a baseline of 3. The description's id guidance largely restates the schema's own 'from a list_feedback result' wording, adding only the emphatic 'do not guess or reconstruct' caution, and it says nothing about the stage/project 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?

States a specific verb and resource ('Replace an existing feedback item with corrected text') and immediately differentiates itself from two siblings by naming them and the conditions that select them. An agent can pick this tool over remove_feedback or record_feedback without opening any schema.

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?

Explicit when-to-use routing: 'To remove an item without replacing it, use remove_feedback; to add new feedback, use record_feedback.' It also gives an in-tool precondition (use a feedback_id from a prior list_feedback result, never guess) rather than leaving it to inference.

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.