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

Annotations declare destructiveHint=true and non-idempotent, but the description is what explains why: the retry costs money like a new run and requires a human approval gate. It also discloses that a workflow retry uses the current definition rather than the original snapshot, a substantive re-execution semantic not present in any structured field.

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 compact sentences ordered by importance: what it does, what it costs and how to gate it, then the diagnostic prerequisite. No filler and no repetition of the title or schema.

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 two-parameter mutation with no output schema, everything needed to call it safely is present: cost awareness, the approval handshake, and the diagnose_run prerequisite. The only implicit element, that the unapproved first call returns a cost estimate, is clearly signposted.

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%, so the baseline is 3, and the description earns a bump by explaining the operating protocol around 'approved' — its first-call/second-call usage and its dependency on a person's consent plus cost — rather than merely restating the schema's phrasing. It adds little about 'run_id' beyond the schema, so it stops short of a 5.

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 and resource ('Start a run again') and immediately scopes it: same inputs for a workflow 'as it is now', same message for an agent. That parenthetical also disambiguates the two execution modes a caller could mean, which the name alone cannot convey.

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?

Gives an explicit sequencing rule — call first without approved to obtain the cost, then call again with approved: true only after human consent — and names a prerequisite alternative ('Check diagnose_run first: some failures need a fix'). It also states the condition under which retrying is the wrong move, which is the rare part.

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