history

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

Well beyond the destructiveHint=true annotation, the description explains that undo is itself recorded ('undoing again redoes it'), that a deleted record is restored, and that the operation is refused when the record changed since the last read. These are exactly the behavioral traits an agent needs before attempting a destructive write.

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 content is dense and mostly earns its place, and the top-level purpose is front-loaded. But it is delivered as one long run-on block with inline 'action=' asides, hurting scannability for a 12-parameter, 3-action tool.

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 and no annotations covering return shape, the description still sketches what comes back ('old and new values', 'newest first, across projects or one project', the app/model/machine fields). That is good coverage for a complex multi-action tool, though pagination behavior and the 'limit' parameter remain underexplained.

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?

With 67% schema coverage, the description adds real meaning: it maps by_app/by_model/by_machine/since/until/page to the 'who' action and gives a concrete time syntax ('2026-10-17 or 2026-10-17 14:30'). It also ties record_type + record_id to single-record listing and states project is 'required for every write'. It leaves 'model' and 'limit' unexplained, which the schema partially covers.

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 description names specific verbs and resources ('Edit history of FlowState records, and undo') and then breaks the tool into three named actions (list/undo/who), each with a distinct read or write purpose. It does not, however, distinguish itself from siblings like get_recent_changes or log_change, which an agent could plausibly confuse with the 'list' and 'who' actions.

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?

Each action is given a clear context ('action=list (read): recent edits and deletes in a project', 'action=undo (write; project required)'), and real exclusions are stated ('New records are not undone here (set them done/cancelled)', 'Refused if the record changed again since'). It never names alternative tools, so an agent has no explicit routing guidance among overlapping siblings.

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