agent-work-mcp

Overview Schema Source Related Servers Discussions Admin

Update work graph

update_work_graph

Add or update work units, cancel or reopen them, and record project context. Enforces valid, non-cyclic dependencies and requires rationale for new units after the initial plan.

Instructions

Add or update units (upsert by id), cancel or reopen units, and record project context (summary, key files, commands). The first call submits the initial plan. Rejected as a whole if ids, references or dependencies are invalid or cyclic. Units added after the initial plan need a rationale. Never add work only to use remaining budget.

Input Schema

TableJSON Schema

Name	Required	Description
`units`	No	Units to add or update (upsert by id).
`cancel`	No	Remove units from scope.
`reopen`	No	Return blocked, failed or cancelled units to pending (resets attempts).
`sessionId`	Yes	Session handle returned by start_session.
`projectContext`	No

Output Schema

TableJSON Schema

Name	Required	Description	Default
`added`	Yes
`state`	Yes
`updated`	Yes
`guidance`	Yes
`reopened`	Yes
`cancelled`	Yes
`sessionId`	Yes
`readyUnitIds`	Yes
`graphRevision`	Yes

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 19, 2026

TDQS

A4.4/5.0

Behavior4/5

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

With annotations carrying only default/negative hints, the description carries the transparency burden. It discloses atomic rejection on invalid/cyclic ids, explains that later-added units require a rationale, and warns against budget padding. It does not detail side effects of cancel on dependents, but the key behavioral traits are surfaced.

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?

Four sentences with no redundancy; the main operations are front-loaded, constraints follow logically, and the final sentence is a crisp boundary. Every sentence adds information.

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?

Given the tool's multi-operation complexity and the presence of an output schema, the description covers the essential usage context, validation behavior, and budget expectations. It omits minor details like dependency-cancellation semantics but is sufficient for correct invocation.

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 high (80%), so the baseline is 3. The description adds beyond the schema by explaining upsert semantics, the cancel/reopen actions, and the initial-plan condition that determines when rationale is required, giving inter-parameter meaning.

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 opens with a specific verb-resource pairing ('Add or update units... cancel or reopen... record project context') that makes the tool's function unmistakable. It also names the distinguishing context ('The first call submits the initial plan'), separating it from read-only siblings like get_work_graph and execution tools like next_work.

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?

It gives clear context on when to use the tool: the first call initializes the plan, and later calls add/update with rationale requirements. It includes a when-not ('Never add work only to use remaining budget'), though it does not explicitly name sibling alternatives.

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

Deploy Server