Gc

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

The description does add real behavioral context (colocated=True reservation, advisory arbitration, 'honest degradation' on dirty trees), which is more than the bare annotations give. However, for a GC operation the critical behaviors — what gets deleted, whether reclamation is reversible, interaction with active leases — are entirely absent, and `destructiveHint: false` is left unexplained against a tool whose params say it 'reclaim[s] cache entries'. No outright contradiction in the description text itself, but the disclosure misses the operation it names.

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 text is long and back-loaded with meta-commentary (CONCEPT:RM-TASK-LEDGER, D-CP-8, bold ALL-CAPS fragments) before anything actionable. Much of it — 21.7 GB measurements, cargo/pnpm anecdotes, NFS lease details — is irrelevant to invoking this specific tool, and the tool's actual action is never front-loaded.

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?

An output schema exists so return values need no explanation, but the description omits the core of a GC tool: what state it mutates, whether it is safe to run concurrently with builds or lanes, and any prerequisites. Given seven optional params with multi-mode behavior, the description leaves the agent without the information needed to call it correctly.

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 100%, so the baseline is 3; each parameter including `keep_recent`, `max_age_days`, and `wait_timeout` is already documented in the schema. The description adds only the conceptual key tuple `(repo, tree-sha, feature-set, toolchain-fingerprint, target-triple)`, which mildly enriches `key` but does not change how any param is passed.

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?

The name and title say 'Gc', and two params are explicitly 'For gc', yet the description never states that this tool reclaims/garbage-collects cache entries. Instead it describes a build broker that 'publishes named artifacts to a shared, checksummed cache' — an operation belonging to sibling tools like rm_build_request. An agent reading only the description would not know this tool's verb.

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?

There is guidance for routing build *requests* through the broker and for the remote `host` path, but nothing on when to run a GC, how it interacts with in-flight builds, or when to prefer it over rm_projects_build/cleanup tools. The guidance present is for a different mode than the tool name implies.

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