update_citation

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

Annotations already declare readOnly=false, destructiveHint=true, idempotentHint=true, so the safety profile is covered and the description is consistent with it. Beyond that, the description discloses a real blast-radius trait: whatever confidence you set applies to EVERY fact attached to the citation, and it explains the per-claim model. It doesn't mention reversibility (undo_transaction exists) or partial-update behavior, so not a full 5.

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 primary verb and affected fields are front-loaded, and the IMPORTANT block is the highest-value content. The middle sentence about page-less citations is somewhat abstract and cryptic, costing a little density, but nothing is truly wasteful.

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 full annotation coverage and no output schema, the description supplies the domain-specific warning an agent needs (one confidence per citation, shared across attached facts) and the partial-update hint ('omit to leave unchanged'). Missing only peripheral details like error/permission behavior, which sibling tools arguably cover.

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%, so the baseline is 3; the description still adds interpretive meaning beyond the schema by framing 'page' as the locator (a page-less citation is not a locator), clarifying that confidence is a per-instance judgement rather than a source-class property, and explaining when to use 'source' (re-pointing a mis-cited or split container). This is genuine semantic value on top of structured fields.

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 resource ('Edit a citation') and enumerates the editable facets (locator, confidence, date, source), so the agent knows exactly what this tool mutates. It does not differentiate itself from the sibling update_citations (batch plural), which is the most obvious confusion risk.

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 gives a strong edge-case rule -- when the same page supports a second distinct claim, create a SECOND citation rather than re-grading this one -- which is actionable guidance. However it never routes the agent between this and the plural update_citations, add_citation, or get_citation siblings, so alternative selection is left to inference.

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