arf_parser

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

Annotations only declare readOnlyHint=true, and the description adds real behavior beyond that: session-level file caching (no re-parse), the interaction with manual exclusions, the shape of the return value (Markdown summary with top positive/negative loadings plus JSON for score plots), and the caveat that subset selections break global normalization. That is a lot of genuine context for a read-only tool.

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?

Purpose and the key alternative come first, then a parameter inventory; length is proportional to 20 arguments and no bullet is pure filler. It is dense rather than padded, though the class_ids/group_factors examples and the TIC caveat could be tightened slightly.

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 read-only, 20-parameter analysis tool with no output schema, the description covers inputs, caching, exclusion interaction, return contents, and downstream implications (re-running preprocessing after spot selection changes). An agent has what it needs to call this correctly.

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% and there are 20 parameters, so the description must carry the load — and it does, documenting defaults and edge semantics for nearly every argument: annotation_keyword vs ontologies (PC matching LPC), class_ids token AND/OR resolution across Class ID and sample name, group_factors vs group_levels precedence, and include_roles defaulting to sample-only. This is well beyond a restatement 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?

States a concrete verb chain (load .arf, apply filters, run PCA) and declares itself the single entry point for PCA on the raw spot matrix. It also explicitly separates itself from arf_pca_preprocessed, which handles the preprocessed matrix, so an agent can route between the two without opening schemas.

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?

Gives explicit when-to-use rules: re-call with different arguments to redo PCA (noting the file is session-cached), and use arf_preprocess → arf_pca_preprocessed for normalized/QC'd matrices. It also notes arf_exclude changes are applied before PCA and warns about TIC normalization on subsets.

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