deep_search

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

With no annotations, the description carries the full burden; it does disclose meaningful behavior — that work is delegated to an external CLI and that the point is context economy. It does not say whether the operation is strictly read-only, what the response looks like (report vs. raw hits), or anything about latency/cost of spawning a CLI agent.

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?

Two sentences, front-loaded with the delegation model, then the concrete capability list, then the selection rule. No filler; even the 'archaeology' metaphor earns its place by framing history-focused search.

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 three-parameter, one-required tool with no annotations and no output schema, the description covers intent and routing well. The notable gap is return shape and safety profile — an agent cannot tell from this text what it gets back or whether the CLI can touch the working tree.

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 cwd, model, and query are all documented in the schema and baseline 3 applies. The description's example phrasings ('when/why did X change', 'where is Y used') usefully signal that query should be natural language, but that meaning is largely already conveyed by the schema's own example.

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-plus-resource ('Delegate codebase archaeology... git log/diff/blame spelunking, wide greps') and concrete example intents, so an agent knows exactly what class of work it covers. It does not name or contrast the sibling tools (analyze_files, web_lookup, delegate), leaving the codebase-vs-file-analysis boundary implicit.

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?

'USE THIS instead of running many search commands yourself — it saves your context' is an explicit selection rule with the alternative it displaces. It stops short of stating when NOT to use it (e.g., a single known file read, or a plain web lookup), so the exclusion side is missing.

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