mark_done

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

The description goes far beyond the annotations by disclosing exact error semantics (404 vs 400), idempotent behavior, time-tracking side effects, pinned-task behavior, and `day`-field handling. It even notes server-behavior changes over time. This is rich behavioral context that the annotations alone could not provide, and it does not contradict the idempotentHint=true or destructiveHint=false annotations.

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 description is information-dense and mostly valuable, but it opens with two overlapping sentences ('Mark a task as done; projects are completed in the app.' followed by 'Mark a task as done (via /markDone...)'). Several live-testing dates and caveats could be consolidated. Still, the length is largely justified by the many behavioral edge cases.

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 one parameter, an output schema, and idempotence annotations, the description covers all decision-relevant context: supported vs unsupported resources, error codes, side effects on time tracking, recurring tasks, pinned tasks, and how to read back completed tasks. Nothing an agent needs to invoke it correctly is missing.

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?

The schema already describes item_id as 'Task ID (NOT a project — see description)', and the description reinforces this with details about deterministic recurring-task instance IDs and the distinction from project IDs. Though schema coverage is 100%, the description adds useful value about what kind of ID is valid and how recurring instances are formed.

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 clearly states a specific action ('Mark a task as done') and resource (tasks), and explicitly distinguishes from projects ('Tasks ONLY: for projects the API responds 400'). It also differentiates from related tools like /doc/update and unmark-related behavior. This leaves no ambiguity about the tool's core purpose.

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?

The description gives explicit when-to-use guidance and names alternatives: projects should be completed in the Marvin app, /doc/update would technically work but skips side effects, and get_done_items should be used for listing completed tasks. It also covers edge cases like recurring-task instances and error conditions, so an agent can decide when to invoke this tool versus others.

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