kuma_safety

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

With no annotations, the description carries the full burden and does reasonably well: it discloses that guard checks anti-patterns/drift/loops, verify runs scoped tests plus validation, and checkpoint/rollback is 'the one rollback mechanism' that snapshots and restores. It stops short of stating side effects (filesystem snapshots), persistence, or permission/caching behavior — the cache/rate-limit hint only surfaces in the schema's force param.

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 purpose sentence, then a compact action list where each parenthetical earns its place by adding timing or output context. Slightly dense but no wasted sentences.

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 an 11-parameter, no-annotation, no-output-schema dispatcher, the description supplies the action routing and timing that the schema cannot. It omits any mention of what the actions return or of side effects such as snapshot creation, which leaves a small but real gap.

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%, including a per-value description on the action enum, so the structured data already documents every parameter. The description's action glosses largely restate the schema's enum description, adding little new syntax or format meaning, so baseline 3 is appropriate.

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 domain (safety & verification) and enumerates the five core actions with a one-phrase purpose each, so an agent knows exactly what this dispatcher does. It does not explicitly differentiate itself from siblings like kuma_context or kuma_memory, though the safety/verification framing implies the split.

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 gives an explicit when ('use at task boundaries') and an explicit when-not ('not on every edit'), plus per-action timing: guard before risky work, verify after edits, checkpoint before risky work, rollback after. This is exactly the routing guidance a multi-action tool needs.

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