Trent

Update remediation tasks

update_tasks

Destructive

Update remediation control statuses in a batch. Updates are atomic and status changes persist across scans.

Use this for user-requested manual overrides. After implementing fixes,
run trigger_analysis so Trent confirms resolved controls. Do not mark
controls completed on your own initiative; doing so overrides verification.
control_status accepts in_review, in_progress, excluded, or completed.
remediation_category switches between automatable and manual.

For a risk decision, use record_feedback to inform the scan or
edit_analysis with field='disposition' on a task. For a finding, use
edit_analysis with reason='risk_accepted'.

completed requires a reason of at least 20 characters. excluded accepts
the control's risk. When the user requested either change, pass
confirmed=true to apply it. If you propose it, show the reason and ask
first. A completed update with confirmed=false returns justifications
for review and leaves controls unchanged.

ticket_action files each control's ticket or posts a note in this call.
Use it when requested by the user; ask first when proposing it. The reply
names each filed ticket with key and link, and each posted note.
tickets_refused explains failures, the remedy, and who can act. Relay its
message verbatim. The retry call targets the failed writes.

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": {
      "updates": "<updates>"
    },
    "name": "update_tasks"
  }
}

Input Schema

Name	Required	Description	Default
`note`	No	With ticket_action='create': an optional comment on each new ticket. With ticket_action='note': the comment to post, required. Trent posts it as the user's comment.
`updates`	Yes	List of control updates to apply.
`confirmed`	No	True when the user requested completed or excluded status, or approved it after reviewing the justifications. False with control_status='completed' returns those justifications and leaves controls unchanged.
`project_id`	No	Project ID (provide either project_id or project_name)
`project_name`	No	Project name (provide either project_id or project_name)
`ticket_action`	No	Ticket action for controls in updates. 'none' (default) leaves tickets unchanged. 'create' files a ticket for each control at any priority. 'note' posts note on each control's ticket. Ticket actions can be taken with control status unchanged.	none

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.
`results`	No	Per-control update results from Trent.
`tickets`	No	The tickets this call filed or posted a note on, one per control: control_id, key, url, status and trent_state. status 'creating' means Trent is still filing the ticket and retries on its own.
`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.
`outcomes`	No	When the call did more than one thing: what each part did, in order. message gives the first.
`next_steps`	No	Recommended next actions, in order.
`retry_controls`	No	Controls whose ticket write failed and can be sent again. After fixing the cause in tickets_refused, retry those controls with the same ticket_action and note. next_steps gives the retry call.
`tickets_refused`	No	Controls whose ticket creation or note failed, each with control_id, code, and a message explaining why and who can fix it. Relay the message as written.
`retry_after_seconds`	No	With reason rate_limited, how many seconds Trent asks the caller to wait.
`pending_completed_controls`	No	Completed controls awaiting explicit user approval.

Schema Changelog

Changes observed during successful MCP inspections.

Changed3 schema fields changedOct 30, 2026

addedOutput schema / properties / outcomes

Added value: +{
+  "anyOf": [
+    {
+      "items": {
+        "type": "string"
+      },
+      "type": "array"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "default": null,
+  "description": "When the call did more than one thing: what each part did, in order. message gives the first.",
+  "title": "Outcomes"
+}

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"
+}

changedOutput schema / properties / retry_controls / description

Previous value: -"Controls whose ticket write failed and can be sent again. After fixing the cause in tickets_refused, retry those controls with the same ticket_action and note. The message gives the retry call."New value: +"Controls whose ticket write failed and can be sent again. After fixing the cause in tickets_refused, retry those controls with the same ticket_action and note. next_steps gives the retry call."

First observedOct 23, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Goes well beyond the destructiveHint annotation: states atomicity, persistence across scans, that a completed update with confirmed=false returns justifications and leaves controls unchanged (a built-in dry-run), and that ticket_action files tickets or posts notes in the same call. It also describes failure handling: tickets_refused explains the failure and remedy, must be relayed verbatim, and the retry targets only the failed writes.

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?

The workflow constraints and alternatives are front-loaded and every paragraph conveys an actionable rule. It is dense and fairly long, with the final ticket/reply paragraph sitting somewhat apart from the core update behavior, but little is redundant.

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?

A destructive, open-world mutation tool with an output schema; the description covers the confirmation model, the review-only path, side-effect ticket behavior, and failure/retry semantics. Nothing an agent needs to invoke it safely is missing.

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 baseline is 3; the description still adds workflow meaning for confirmed (gates completed/excluded, triggers a justification-only response) and ticket_action (per-control ticket creation or note posting, allowed with status unchanged). The enum values for control_status and remediation_category are mostly restated from the schema rather than extended.

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 first sentence names the specific verb, resource, and scope: batch updates of remediation control statuses with atomic application. It also states a distinguishing property (changes persist across scans) that separates it from analysis-editing siblings.

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?

Explicitly routes between tools: record_feedback for a risk decision, edit_analysis with field='disposition' for a task, edit_analysis with reason='risk_accepted' for a finding, and trigger_analysis after fixes are implemented. It also states a negative constraint ('do not mark controls completed on your own initiative') and the confirmed=true vs. ask-first rule.

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.