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

Annotations declare the mutation profile (destructive, idempotent, non-readonly), and the description adds substantial context beyond them: the operation is atomic, Brain Scanner computes sizes and digests internally, and a concrete rejection list (source contents, secrets, commands, diffs, logs, environment values, credentials). It also discloses that narrative prose is never generated by the server, which is non-obvious agent-relevant behavior.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Is the description appropriately sized, front-loaded, and free of redundancy?

Three dense sentences, front-loaded with the action and its scope, then constraints, then the narrative obligation. Every sentence carries information, though the phrasing is packed enough to read as jargon-heavy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a destructive, high-cardinality mutation with no output schema, the description covers atomicity, privacy filtering, and the narrative contract. It omits failure/conflict behavior and the practical graph size limits, but expectedProjectVersion's schema description carries the concurrency guard, so the remaining gap is minor.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 50%, with the covered parameters (projectId, sessionId, idempotencyKey, expectedProjectVersion) already well documented in-schema. The description adds meaning only for changeNarrative (four plain-language sections, manually authored), leaving nodes/edges semantics and the size bounds to the schema. Baseline 3 is appropriate given partial compensation.

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

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

States a specific verb (publish) and resource (project graph) with scope qualifiers: 'atomically', 'one bounded, privacy-filtered'. The phrase 'Preferred agent path' implicitly distinguishes it from siblings like patch_project_graph, export_project_graph, and validate_project_graph, so an agent can route without opening schemas.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Does the description explain when to use this tool, when not to, or what alternatives exist?

The 'preferred agent path' framing gives clear context that this is the sanctioned publishing route, and the description states a concrete usage requirement (the agent must write the four changeNarrative sections itself). It stops short of naming alternatives or stating when-not to use it, e.g. incremental updates via patch_project_graph.

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