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

The description says 'READ-ONLY: mutations refused', but the annotations explicitly set readOnlyHint=false. This is a direct contradiction about whether the tool performs a read-only operation or executes a mutation. The description also adds sandbox and notarization concepts, but the fundamental safety profile is contradictory.

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 short and front-loaded with the main action, which is good. However, the parenthetical creates structural confusion by placing a contradictory constraint after the main clause rather than clarifying whether the tool executes or refuses mutations.

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 output schema, the description should clarify what 'notarize with Spec 0.5 receipt' returns and what an agent should expect after evaluation. It also leaves the read-only contradiction unresolved, so the agent lacks enough context to invoke this tool confidently.

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 100%, so candidate_id, command, and mutation_kind are already fully documented by the schema. The description does not add any further meaning about parameter interactions, defaults, or enum behavior beyond what the schema provides.

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 names a specific action (execute a candidate mutation) and sandbox/notarization context, but the parenthetical 'READ-ONLY: mutations refused' directly conflicts with the apparent purpose of executing mutations, making the tool's actual role ambiguous. It also does not distinguish mandala.evaluate from siblings like mandala.triage or mandala.status.

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?

There is no guidance on when to use this tool versus alternatives such as mandala.triage, mandala.status, or receipt tools. The parenthetical hints at a restriction, but it does not state the conditions for invoking this tool or what to do instead when mutations are refused.

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