reconcile_tasks

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

No annotations provided, so description fully carries burden. It discloses that safe_fix writes state transitions atomically with backup (status.json.bak), audit fields, and reconcile.log. Also states it never touches tasks owned by an active watcher, and only applies high-confidence fixes.

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?

Description is somewhat long but every sentence provides essential information. Front-loaded with version and core purpose. Could be slightly tightened but no wasted words.

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?

No output schema, but description indicates return types (diagnosis report, status transitions) adequately. Covers all parameters, side effects, and constraints. Clearly differentiates from many sibling task tools.

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 100%, but description adds significant meaning: clarifies 'report_only' vs 'safe_fix' behavior, defines 'max_age_minutes' as based on created_at or mtime, and explains 'include_done_candidates' for auditing. This goes well beyond the schema descriptions.

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 the tool scans stale running/collecting_artifacts tasks and either reports or safely fixes them, using specific verbs and resource. It distinguishes the two modes (report_only vs safe_fix) and mentions constraints like not touching active watcher tasks or low-confidence fixes.

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?

Explicitly explains when to use each mode: report_only for diagnosis without modification, safe_fix for applying high-confidence transitions with backup and atomicity. Also states what it never does (medium/low confidence, active watcher tasks), guiding appropriate use.

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