knowledge_admin

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

With no annotations provided, the description carries the full burden of behavioral disclosure, and it does disclose destructive effects for prune_orphans ('delete embeddings') and vacuum ('reclaim free pages'). However, it understates rebuild_embeddings' destructive rewriting of existing embeddings, and it completely omits the 'promote' action's behavior (apply vs explain modes, write behavior vs read-only). For a tool with several destructive operations, the disclosure is partial.

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 core operations are front-loaded in the opening sentence, with each action's purpose expanded in a single subsequent sentence. The structure is efficient and scannable. Minor deduction for the slightly run-on single-sentence expansion and for leaving out promote, which makes the list feel incomplete rather than intentionally concise.

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 complex admin tool with 10 parameters, 6 actions, no output schema, and no annotations, the description should be comprehensive — but it covers only 5 of 6 actions and omits the entire promote workflow with its 4 dedicated parameters (promote_mode, min_score, min_recall_count, min_unique_queries). An agent reading only the description would never learn a whole callable mode exists, which is a significant completeness gap.

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 schema already documents all 10 parameters, giving a baseline of 3. The description adds marginal value by linking rebuild_embeddings to the provider-switching use case, but it adds no semantics beyond the schema for the config params (git_url, memory_dir, auto_distill) and leaves the promote-related parameters entirely unexplained in prose despite them being a substantial cluster of the schema.

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 states a clear purpose — administrative operations on the knowledge/vector store — and enumerates five concrete actions (status, config, rebuild_embeddings, prune_orphans, vacuum) with their effects. It clearly distinguishes itself from the read/search siblings. However, it silently omits the 'promote' action that exists in the schema enum, and the umbrella phrase 'Admin operations' is vague until the action list clarifies it.

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?

The description gives actionable when-to-use context for rebuild_embeddings ('useful when switching providers') and explains what each action accomplishes, which implicitly routes the agent to the right action. But it provides no guidance for the 'promote' action (its dedicated params min_score, min_recall_count, min_unique_queries, promote_mode are never narrated), and it never states when NOT to use the tool or how it differs from siblings beyond 'admin' framing.

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