preprocess_data

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

Annotations indicate destructiveHint=true, but the description does not elaborate on what is destroyed or modified (e.g., whether it overwrites existing data, irreversibility, or side effects). The only behavioral note is a negative scope (not computing embeddings), which is more purpose-related than a behavioral disclosure. With a destructive hint, the description should warn about data overwrite or similar consequences.

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 concise and well-structured: it opens with the primary purpose, then clarifies scope limits, and ends with a compact Args section. Every sentence adds value without redundancy, making it easy for an agent to quickly grasp the tool's core function.

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's complexity (many parameters, destructive behavior) and the lack of output schema details, the description is somewhat minimal. It covers purpose and the boundary with compute_embeddings, but it omits prerequisites (e.g., data must already be loaded), side effects of the destructive operation, and what the function returns. This is adequate but with clear gaps.

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?

The input schema itself provides detailed descriptions for each parameter, so the baseline is 3. The description adds only minimal value by saying 'all have sensible defaults,' which helps convey that params can be omitted, but it does not explain any parameter semantics beyond that. The schema carries the heavy lifting for parameter understanding.

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 states the tool's function with a specific verb and resource: 'Run QC, filtering, normalization, and highly variable gene selection.' It also explicitly differentiates itself from a sibling tool by stating it does not compute PCA, UMAP, clustering, or neighbor graphs, and directs users to compute_embeddings() when those artifacts are 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 clear context by stating that compute_embeddings() should be run afterward when downstream tools require those artifacts. This implies the tool's role in the analysis pipeline, but it does not explicitly mention when not to use this tool or list alternative preprocessors, so there is room for improvement.

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