Build an evaluation test set from real conversations

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 that it is read-only apart from writing a file, and that it reads sessions from a source. It explains the 'onlyFailed' behavior and hints at portal import. However, it doesn't disclose details like rate limits, authentication requirements, or potential side effects beyond file creation. The description is fairly transparent but misses some depth that an agent might need.

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, front-loaded with the primary purpose and the key behavioral distinction. It packs a lot of information into a compact space without redundancy. Every sentence adds something: purpose, the onlyFailed behavior, and the usage context.

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 10 parameters with full schema coverage, an output file that is the CSV, and no output schema, the description provides enough context: what it does, how to use it, and the next step. It doesn't detail the CSV format or error handling, but those are not required for an agent to invoke it correctly. The description is complete for practical use.

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% with detailed parameter descriptions, so the baseline is 3. The description adds value by explaining the purpose of 'onlyFailed' in context and clarifies that the output file is the import CSV. It also mentions 'real conversations' which aligns with top/days parameters. However, it doesn't add much beyond the schema for parameters like botId or environmentId, but the schema already explains defaults clearly.

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?

Description clearly states what the tool does: writes a CSV from real user questions, ordered by frequency. It distinguishes itself from guessing from workspace content and mentions 'onlyFailed' filtering. It also names a sibling tool (cs_run_evaluation) for the next step, differentiating its role.

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?

Description explains when to use it: when you need a test set derived from actual conversations instead of guesses. It mentions the 'onlyFailed' flag for selecting valuable sessions and notes the workflow: import once in portal, then use cs_run_evaluation. It doesn't explicitly say when NOT to use it (e.g., if you need all sessions or synthetic data), but the guidance is clear enough.

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