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

Annotations already declare non-readonly, idempotent, non-destructive. The description adds real context beyond that: edits are gated by project mode, only slots without a video are editable, and facts must come from the user rather than being invented. That last constraint is an important behavioral rule no annotation encodes. Missing is what happens to an existing approval or script after an edit.

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 sentences, front-loaded with the action, then the two mode branches, then the follow-up call. The parenthetical '(the warning says so)' is a small indulgence but ties the rule to something the agent will observe.

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 four-parameter mutation tool with no output schema, the description covers the important ambiguity: which slots are editable per mode and what facts are acceptable. No return-value explanation is needed since there is no output schema. It leaves open the post-edit state (whether approval is cleared).

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 only 25% (only 'facts' is documented in the schema), so the description must compensate. It does name idea, angle and facts as the editable fields and reinforces the facts semantics ('add the facts the user gives you, never invented ones'). But it adds nothing about slot_id or the meaning of angle vs idea beyond their names, so compensation is partial.

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 target with the editable fields enumerated: 'Changes a slot's idea, angle and facts before its video is made.' An agent immediately knows the mutation. It does not, however, distinguish itself from the sibling change_slot_idea, which appears to touch the same 'idea' field, so sibling differentiation is missing.

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 mode-dependent conditions: in Automatic only slots needing real facts are editable, in Manual any idea or approved slot without a video. It also routes to the next step ('then approve_slot'), which is genuinely useful chaining guidance. It stops short of naming change_slot_idea as the alternative for pure idea edits.

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