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

Annotations declare it writable, destructive and idempotent, but the description adds material context the annotations can't convey: nothing happens until a human approves the proposal in Chief, queue_now requires the milestones:approve permission and user confirmation, and duplicates are refused with a specific duplicate_of error unless allow_duplicate is true. What's missing is the return shape or what state the milestone enters after approval, which keeps it from a 5.

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?

Six tight sentences in priority order: purpose, default state, conditional flag, content requirements, duplicate avoidance, retry key. Nothing is repeated from the schema and every sentence carries an operational instruction.

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 an 8-parameter, 5-required mutation tool with no output schema, the description covers workflow, permissions, deduplication and idempotency well. It stops short of describing what the call returns (e.g., the proposal id) or limitations such as rate limits, so it is strong but not exhaustive.

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%, yet the description still adds meaning for the two non-obvious booleans (queue_now's confirmation and permission requirement, allow_duplicate's role) and for idempotency_key's retry semantics. It also characterizes the required content fields ('clear objective and a checklist definition of done'), though project/title/context get no extra guidance.

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?

Opens with a specific verb+resource ('Propose a new milestone in a Chief project') and immediately states the default state (created as a PROPOSAL). This clearly separates it from siblings like approve_milestone, deny_milestone, or list_milestones, which the agent can distinguish without opening a schema.

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 when-to-use rules for the ambiguous parameter: 'Set queue_now only when the user asked for this task to be queued'. It also names the prerequisite sibling action ('Call list_milestones first to avoid duplicates') and prescribes retry behavior ('Reuse the same idempotency_key when retrying'). Alternatives and exclusions are all spelled out.

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