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

Annotations flag the whole tool destructive, but the description goes further by localizing the destructiveness to restore and revert_prompt, and clarifying that revert_prompt touches 'the prompt only' while restore 'replace[s] the agent'. It adds a user-confirmation requirement and notes the performance/cost effect of set_prompt_caching, none of which the annotations convey.

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?

Risk guidance and the update_agent cross-reference are front-loaded, and every action entry carries distinct information with no filler. The pipe-delimited action list with trailing 'Needs: ...' fragments reads as a run-on and slightly hurts parseability, which keeps it from a 5.

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 6-parameter, no-output-schema mutation tool, the description covers all actions, their required inputs, side effects, and sequencing. The one real gap is that it never says what 'create' returns (presumably a backupId) or how the agent learns it for a later restore, which is the natural next step in this workflow.

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?

The action parameter has no schema description at 67% overall coverage, and the description fully compensates by pairing each enum value with its meaning and required companion parameter ('restore ... Needs: backupId', 'revert_prompt ... Needs: snapshotId', 'set_prompt_caching ... Needs: enabled'). It adds provenance the schema lacks, telling the agent where backupId and snapshotId come from.

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 opening sentence gives a specific verb pair and resource ('Back up or roll back an agent'), then enumerates all four actions with one-line meanings, so an agent can distinguish create/restore/revert_prompt/set_prompt_caching without opening the enum. It also implicitly separates itself from the read-side sibling by pointing backupId at 'agent_backups_read list' and snapshotId at 'prompt_history'.

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?

Explicit sequencing guidance: 'Take a backup (create) BEFORE large edits with update_agent' names the sibling and the trigger condition, and 'restore and revert_prompt overwrite the current agent — confirm with the user and create a backup first' states both the risk and the required prerequisite. This is genuine when-to-use and when-to-be-careful guidance rather than implied usage.

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