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

The description adds substantial context beyond the annotations: retention of unchanged constraints, a 24 h expiry on references, and a privacy disclosure that Agentery sees only the request and correction, never the wider conversation. It does not, however, address the destructiveHint=true / idempotentHint=false profile, so an agent gets no confirmation that re-running mutates remembered task state rather than user data.

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?

It is one dense paragraph with the primary trigger front-loaded, which is good. But the source enum is stated twice ('source: human, agent, follow_up' and later 'source is one of human, agent, follow_up'), and several trailing sentences describe how sibling task answers are formatted to be answered, which is not this tool's input contract.

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 stateful refinement tool with no output schema and only three parameters, the description covers what is returned ('the new answer plus what changed'), the reference lifecycle, and the privacy boundary, which is what an agent needs. Minor gap: nothing on failure modes or what happens if the request_ref has expired.

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 already 100%, so the baseline is 3. The description exceeds that by giving concrete correction examples ('must integrate with Slack', 'budget is $20/month'), explaining the request_ref origin, and disambiguating the source enum values ('the user's words' vs 'your assessment, default' vs 'follow_up').

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 gives a specific verb and resource: it re-runs an earlier search with changed constraints, keeps the others, and returns the new answer plus what changed. It names the origin tools (search_providers, research_capability) and the task-tool family, so an agent can place it among siblings. The core 'what it does' is stated mid-sentence rather than up front, which costs a point.

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?

It opens with an explicit trigger ('Use when an earlier search_providers or research_capability answer missed what was meant') and tells the agent exactly what to pass. It also routes the alternative case ('Pass the same request_ref to another task tool to continue the task there'), so both when-to-use and when-to-use-something-else are covered.

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