tarn_last_failures

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

With no annotations provided, the description carries the full burden of disclosing behavioral traits. It clearly states that the tool 'Reads the persisted `failures.json` rather than re-running the tests,' implying a read-only, non-destructive operation. This is sufficient for a simple read tool. However, it does not explicitly mention error handling (e.g., if the file is missing) or authorization requirements, which would elevate transparency further.

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 extremely concise, consisting of just two sentences. The first sentence immediately states the tool's purpose and output format, while the second adds behavioral context and a use case. Every sentence is meaningful and earns its place, with no redundancy or unnecessary detail.

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 that there is no output schema, the description appropriately describes the output as 'structured JSON' of 'grouped failures (NAZ-402).' It also mentions the filename and the read behavior. While it does not detail the exact structure or error scenarios, it is adequate for a simple tool that reads a specific file. The context is reasonably complete for an agent to understand what to expect.

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 already provides 100% coverage for both parameters, including descriptions of defaults and accepted values. The description adds no new parameter semantics beyond what is stated in the schema; it only reiterates the information. Therefore, the description does not compensate or add value beyond the structured data, resulting in a baseline score of 3.

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 'Return the grouped failures (NAZ-402) for a specific run as structured JSON.' It specifies the resource (grouped failures for a run) and the action (return as JSON). Additionally, it distinguishes from siblings by noting it reads a persisted file rather than re-running tests, and highlights a specific use case ('failures-only view without loading the full report'), effectively differentiating from tools like tarn_report or tarn_last_root_causes.

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 guidance on when to use this tool: when you want a cached failures-only view without re-running tests. It implicitly suggests alternatives by contrasting with re-running tests (via tarn_run or tarn_rerun_failed) and mentions 'without loading the full report' as an alternative to tarn_report. However, it does not explicitly list all circumstances where the tool should or should not be used, and lacks explicit exclusions.

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