Recipe drift

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

Annotations already declare readOnlyHint, idempotentHint, and openWorldHint, so the safety profile is covered. The description adds genuinely new behavioral context: it clarifies that despite returning suggestedFixes it never applies them ('Mushi never runs DDL or edits workflows'), and it explains the semantics of the gate status (never_run means not checked, never healthy).

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?

The purpose and return structure are front-loaded, and the long enumeration earns its space because there is no output schema to carry it. It is dense and somewhat run-on, but not padded with filler.

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?

With no output schema, the description carries the full burden and delivers: it spells out the return shape ({ projectId, gates, findings }), the field-level meaning of findings and gate status, and the follow-up action for applying fixes. An agent has everything needed to call and interpret it.

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 coverage is 100% with a single optional projectId, so the schema already documents the parameter fully (including the project_id alias). The description adds no parameter-level meaning beyond that, so the baseline 3 applies.

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+resource ('Return what drifted from the app recipe') and then enumerates the exact drift categories it detects (CI workflows, deploys, env names, migrations, off-token design values). This clearly separates it from siblings like get_app_recipe (which returns the recipe itself) and get_design_deviance.

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?

Usage is implied by the rich enumeration of what it detects — you call it to find drift and get fixes — but the description never names an alternative (e.g. get_app_recipe, list_gate_findings, get_design_deviance) or states when-not to use it. The trailing note on applying fixes hints at a boundary but does not route between tools.

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