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

With no annotations, the description carries the full burden and discharges most of it: fixed routing policy, deterministic moat, unroutable claims quarantined (held, never judged), full set sealed with ordered trace, and input validation. It does not state whether the operation mutates the committed set or is idempotent, leaving a small but real ambiguity.

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?

Front-loaded with the step marker and core action, and dense rather than padded. It is a long single paragraph with several parenthetical asides and heavy capitalization, which slightly dilutes scannability but most sentences carry load-bearing information.

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 complex nested-input, no-output-schema tool with no annotations, it covers invocation constraints, routing behavior, and sealing outcomes well. Return values need not be explained absent an output schema, so the only gap is the read-vs-write nature of the operation.

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 a nested object, so the schema already documents the structure. The description adds genuine meaning beyond it by constraining provenance ('verbatim from candidate_commit') and warning that altered sets are rejected — above the baseline 3 for high-coverage schemas.

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 specific verb ('narrow') and resource ('a COMMITTED set'), and positions itself as 'Candidate Engine step 2', distinguishing it from candidate_commit (step 1) and candidate_get. An agent can place it in the pipeline without opening the schema.

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?

Explicitly names the predecessor ('Pass the set exactly as candidate_commit returned it') and the refusal condition ('Refuses an uncommitted or altered set'), which functions as a clear when-not. The 'never your choice' line tells the agent what it must NOT attempt (verifier selection).

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