holonovel

Agent Task Management

manage_agent

Destructive

Create, advance, settle, and inspect durable NPC/agent tasks and their action lifecycle so GMs can track progress and terminal states.

Instructions

Manage durable NPC/agent tasks and their action lifecycle. create reads goal and optional source_goal; advance appends description; complete, fail, and cancel settle the task and terminal tasks are immutable; autonomy defaults to TTRPG_AGENT_AUTONOMY. Mutating actions persist to the Novel and are audited; list/get are read-only. Use when: creating, starting, advancing, settling, or inspecting tasks. Do NOT use when: surfacing an advisory suggestion — use manage_scene (action: oracle); managing NPC identity — use manage_npc. Revert with manage_history (action: undo).

Input Schema

Name	Required	Description
`goal`	No	Task goal text (create).
`action`	Yes	create, list, get, start, advance, complete, fail, or cancel.
`detail`	No	Return full agent tasks instead of summary entries (list).
`subject`	No	Entity or NPC that owns the task (create/list).
`task_id`	No	Task id (get/start/advance/complete/fail/cancel).
`autonomy`	No	advisory, prompt, or auto (create; default the TTRPG_AGENT_AUTONOMY setting).
`description`	No	Action description (advance).
`source_goal`	No	Originating goal-pursuit suggestion (create).

Output Schema

Name	Required	Description	Default
`text`	No	Human-readable text envelope mirroring the structured result (REQ-001, REQ-548a).
`status`	Yes	Machine-readable result status: OK, NEED_INPUT, WARNING, PARTIAL, or an error category (REQ-002, REQ-548c).

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedv2026.10.13Oct 25, 2026

addedInput schema / properties / detail

Added value: +{
+  "description": "Return full agent tasks instead of summary entries (list).",
+  "type": "boolean"
+}

Addedv1.0.3Sep 29, 2026

TDQS

A4.5/5.0

Behavior4/5

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

Annotations already declare destructive and non-idempotent behavior, but the description adds that mutating actions persist to the Novel and are audited, that terminal tasks are immutable, and that list/get are read-only. It also gives the revert path via manage_history. It stops short of detailing permissions or the exact effects of cancel/fail on stored data, so 4 rather than 5.

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?

Dense but front-loaded; every clause carries routing, lifecycle, or persistence information with no filler. It could be split for readability, but nothing is wasted.

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?

With an output schema present, return values need not be explained. For a complex 8-action mutation tool, the description covers lifecycle, defaults, persistence, audit, usage, and alternatives completely enough to call it correctly.

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 coverage is 100%, so the schema already documents each parameter's associated action. The description adds a useful 'advance appends description' semantic and reiterates the autonomy default, but most of its create/source_goal mapping duplicates schema parentheticals.

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?

Names a specific verb (manage) and resource (durable NPC/agent tasks) and scopes it to the action lifecycle, making it clearly distinct from siblings like manage_npc and manage_scene. An agent can identify the tool's domain without opening the 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?

Provides explicit 'Use when' and 'Do NOT use when' clauses with named alternatives and action paths (manage_scene action: oracle, manage_npc, manage_history undo). This is exactly the routing guidance needed to choose this tool over its siblings.

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