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

Annotations declare a non-read-only, non-idempotent, non-destructive operation, and the description adds real context: mutations are confined to throwaway branches, expensive branch creation is implied, and an apply token is returned only when the baseline is safe and no mutant slipped through. It also flags a human-in-the-loop requirement, which annotations cannot convey and which is consistent with destructiveHint=false.

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?

Three sentences, front-loaded with the action and its guarantees, then the return contract, then the usage instruction. The parenthetical list of mutation examples is dense but each item earns its place by telling the agent what is being tested.

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 single-parameter tool with no output schema, the description covers what the harness does, the gating return value (apply token), and the sequencing relative to schema.apply and branch merges. It omits any indication of runtime/cost despite spinning up multiple branches, and leaves the 'source' parameter undefined, which keeps it short of complete.

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?

There is a single parameter whose schema description ('Schema source') is nominal 100% coverage but semantically thin; the tool description adds no clarification of whether 'source' is a branch id, connection string, or schema file. With high coverage the baseline of 3 applies, but the ambiguity around the required parameter is a real gap.

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 uses a specific verb ('Run the Aegis mutation harness') on a concrete resource and explains the mechanism: prove isolation, then deliberately break each guarantee (FORCE RLS removed, USING (true), WITH CHECK dropped) and require every break to be caught. This is far more than a restatement of the name and lets an agent distinguish it from policy.simulate or schema.apply.

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?

It states explicitly when to invoke it ('before schema.apply or merging a preview branch') and adds the precondition of confirming with the user first. It does not mention policy.simulate, the nearest sibling for policy validation, nor state when it should be skipped (e.g. trivial policy changes), so the alternative-selection guidance is clear but incomplete.

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