tandem_audit

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

With no annotations, the description carries the full behavioral burden and does so thoroughly: sending requires operator opt-in and an OpenRouter key, preview avoids network calls, expected_input_sha256 refuses changed input, repeated pages reuse saved attempts, and there is no automatic retry. It also warns about sensitive report prose and states that advice is not verification. This is exceptionally transparent.

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 front-loaded with the purpose and then groups cost, privacy, and caveat information into distinct, non-redundant sentences. It is long, but every sentence adds a meaningful behavioral fact, so the length is justified for a sensitive, externally-incurring tool.

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 tool with no annotations and significant external side effects, the description covers prerequisites, preview behavior, request contents, privacy exclusions, cost/caching behavior, and retry policy. An output schema exists, so return-value details are not necessary. There is no meaningful missing information that would prevent an agent from invoking the tool 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 description coverage is 0%, and the description does compensate for the two least obvious parameters: preview=true is explained as a bounded export without key/network/reservation, and expected_input_sha256 is explained as a guard against changed input. However, limit, offset, and task_id are left to inference from names and types, and the request-content sentence does not clearly map to specific parameters.

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 opening sentence names a specific action ('explicitly ask Jev'), the target (Jev), and the trigger (gaps in a completed task's reported evidence), so an agent can tell what the tool does. It does not explicitly contrast itself with sibling tools like tandem_review, tandem_findings, or tandem_diagnose, so it stops short of full sibling differentiation.

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 gives clear context for when to call it ('Explicitly ask Jev about gaps...') and provides an exclusion: 'Ordinary results and waiting never call Jev.' It also explains the preview path for inspecting without spending a key. However, it never names a specific alternative tool to use instead, so it lacks the explicit when-not/alternatives mapping of a 5.

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