kb_health_check

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

With no annotations, the description carries the safety burden itself, and it does disclose the consequential behavior: fix=True indexes missing files and rewrites INDEX.md. It also bounds the mutation by saying the inbox backlog is not touched. It omits idempotency and whether rewrite is destructive to custom INDEX.md content.

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 plain-English summary is front-loaded, and the second paragraph's parenthetical labels (backlog, orphans, dead links) compress four report categories into one sentence without wasted words. The closing note about the inbox earns its place by clarifying scope.

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?

An output schema exists, so return formatting needn't be explained, and the description covers the audit categories plus the fix-mode lifecycle (report, repair, report again). The main remaining gap is max_items and whether repeated fix runs are safe.

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 0%, so the description must define both parameters. It explains fix=True well (indexes missing files, rewrites INDEX.md, re-reports), but max_items is never mentioned, leaving the agent to guess whether it caps report entries per category or overall.

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 opens with a specific verb and resource ("Audit the knowledge base") and enumerates exactly what the audit reports: inbox backlog, orphans, dead links, and unindexed text files. That level of detail makes it distinguishable from kb_status and kb_reindex, though it never names those siblings explicitly.

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 clearly states the conditional mode ("with fix=True also repair what it can") and what fix does versus what it deliberately leaves alone ("The inbox backlog is left for you to organise"). It does not, however, name an alternative tool for the unfixed inbox work (e.g. kb_organize or kb_scan_inbox).

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