render_report

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

The annotations already declare openWorldHint=false and destructiveHint=false; the description usefully reinforces this with 'does not send data or open a browser' and clarifies the return contract ('Returns paths'). It also discloses an important behavioral rule beyond the schema: computed facts and ordering remain authoritative, so assistant commentary cannot override them. It does not say whether existing report files are overwritten, which would be the remaining gap.

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?

Four tight sentences, front-loaded with the output artifact and destination before the optional-commentary rule and the no-network guarantee. Nearly every clause carries information; only minor tightening is possible.

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?

There is no output schema, but 'Returns paths' gives an adequate high-level return picture, and the local-only behavior is covered. Against zero schema description coverage and a nested narrative object, the description leaves siteName/clientName semantics and overwrite behavior undocumented, so an agent can call it but lacks full context.

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 description coverage is 0% across 4 parameters (one nested), so the description must compensate. It partially does: 'Optional assistant commentary must reference existing finding IDs' clarifies the narrative.findings[].findingId linkage, and 'inside this audit's local report directory' clarifies auditId. But siteName and clientName are never explained, so coverage is incomplete.

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?

Specific verb (Generate) plus concrete resource (offline HTML action plan and CSV exports) and a defined destination (this audit's local report directory). This clearly distinguishes it from the audit-lifecycle siblings (start_audit, list_findings, audit_status), which retrieve or mutate audit state rather than produce report artifacts.

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?

Usage is implied by the auditId dependency and the 'local report directory' framing, and the description constrains the optional narrative ('must reference existing finding IDs'). However, it never states when this should be called versus other tools, nor any prerequisite such as the audit needing completed findings first.

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