search_files

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

No annotations are provided, so the description carries the full burden. It discloses key behaviors: executes searches in allowed directories, supports content preprocessing (exclude comments/string literals), and offers multiple output formats with sorting/grouping. It also notes the default path behavior. While it does not explicitly state read-only safety, the nature of a search tool implies no side effects, and the description covers most operational aspects. A minor gap is the lack of mention about performance or rate limits.

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 description is three sentences and efficiently front-loads the key purpose in the first two sentences. The third sentence breaks into multiple clauses covering defaults, advanced options, and output flexibility. There is minor redundancy ('advanced... with comprehensive filtering' and then listing filters), but overall every sentence provides distinct information. Could be slightly tighter, but it remains clear and functionally complete.

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?

Given the tool has 24 parameters and no output schema or annotations, the description provides a solid overview. It covers default behavior, supported filters, output formats, and special features like comment exclusion. It does not mention the 'allowed directories' concept or how to discover them (a sibling tool list_allowed_directories exists but is not referenced). Also, 'common patterns' is vague. However, the schema fills many details, and the description is sufficient for an agent to understand the tool's role and basic usage.

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 100%, so every parameter already has a description. The tool description adds value by contextualizing groups of parameters: it lists regex, file type, size, date filtering, and content preprocessing as groups, and explains that excludeComments/excludeStrings are for 'cleaner code searches'. It also clarifies the default search behavior when no pattern is specified – information not explicitly in the schema. This enhances understanding beyond what the schema alone provides.

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 clearly identifies the tool as an advanced file and code search tool with comprehensive filtering. It specifies the verb 'searches for patterns in files' and the resource 'files within allowed directories'. It also lists key capabilities like regex, file type filtering, size constraints, date filtering, and content preprocessing, making the purpose unmistakable. The sibling tools are all different in nature (project exploration, file operations), so no additional differentiation needed.

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 provides useful context for when to use the tool: for pattern-based file searches with many filtering options. It mentions the default behavior when called without arguments ('searches for common patterns in the current directory'), which helps the agent decide when simple invocation suffices. However, it does not explicitly state when not to use it or compare it to alternatives, though sibling tools are sufficiently distinct.

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