test_tuning

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

No annotations exist, so the description carries the full burden. It does disclose valuable traits: the change is applied to a copy (live ruleset untouched), the pcap is re-run (costly operation), before/after per-SID counts are returned, and a decision record is persisted. It omits failure modes, whether before_run_id must reference an existing run, and whether repeated calls are idempotent.

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 sentence is front-loaded, followed by the variant specs and a single worked example that exercises all three parameters. No filler; every line is load-bearing for correct invocation.

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 a nested object with zero schema descriptions, no annotations, and no output schema, the description covers the hard part (change variants) and states the return shape. The remaining gap is the provenance of before_run_id and what happens when the change is invalid or the run id is unknown.

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 the `change` object is opaque (additionalProperties: true), so the description must compensate — and it does, enumerating the exact required fields for all four change.kind variants. before_run_id and reason are only illustrated by the trailing example, not defined (e.g. where the run id comes from, why reason is required).

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 precise action chain: apply one structured change to a copy, re-run the same pcap, return before/after counts per SID. The "to a copy" phrasing implicitly separates it from write_rules_file, but no sibling is named explicitly, so the differentiation is left to inference.

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 implies a hypothesis-testing workflow (test a suppression/filter/rule edit before committing) and says it "Records a decision", but never states when to prefer this over run_pcap, diff_runs, or write_rules_file. Usage is inferable from the workflow rather than declared.

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