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

Annotations only declare readOnly=false, idempotent=true, destructive=false. The description adds substantial behavior beyond that: nothing restarts or is billed on store, secret-looking names or values are refused, and the change only takes effect on the NEXT build. It also warns that the follow-up rebuild replaces the running server, which is critical operational context.

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?

Front-loaded with the core definition before branching into routing and sequencing rules, and every clause carries operational weight. It is dense with parenthetical asides, which slightly hurts scannability, but nothing is padded.

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 mutation tool with no output schema, it covers the safety profile (refusals), the side-effect profile (no restart/billing), the read path (call with neither to retrieve), and the required follow-up step. An agent has everything needed to invoke it correctly and responsibly.

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 75%, and the description closes the remaining gap by explaining the control-flow of the parameters ('vars sets, unset removes; with neither, the current set comes back'), including the get/read behavior that the schema does not express. It also restates the public-prefix constraint that governs valid var names, though projectId and service add little beyond the schema.

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 and resource ('Store PUBLIC build settings') and immediately defines what 'PUBLIC' means in operational terms (values a bundler bakes into shipped JavaScript). It explicitly positions itself against siblings by naming set_compute_secret and env_link as the destinations for non-public data, so an agent can route correctly without opening other 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?

Gives explicit routing rules: secrets go to lavela.set_compute_secret, a whole .env goes to lavela.env_link, and a rebuild must follow via lavela.compute_rebuild (with the caveat to confirm first because it replaces the running server). This is when-to-use, when-not-to-use, and downstream sequencing all in one.

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