Archive agent

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

The description is exceptionally rich about destroy-vs-retain semantics (what is torn down, what survives, no-op on already-archived), which goes well beyond the annotations. However, it contradicts the annotations on idempotency: it states that archiving an already-archived agent 'is a no-op that returns the agent unchanged', which is exactly the idempotent behavior that idempotentHint=false denies. The rest (destructiveHint=true, readOnlyHint=false, openWorldHint=false) is consistent, so this is a partial rather than total contradiction, but it is a real inconsistency an agent could act on.

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 definition is front-loaded with the verb and the full effect list, followed by retention guarantees and then the idempotent no-op case, in a deliberate order. Despite its length, every clause carries distinct operational information (nothing is restated from the name, title, or schema), so no sentence is wasted.

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 destructive, three-id mutation tool with no output schema, the description covers effects, retention, listing visibility, and repeat-call behavior – enough to call it safely. Remaining gaps are modest: no indication of required authorization/grants, or what happens if runtime cancellation is asynchronous or fails partway through the teardown sequence.

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?

All three parameters (orgID, agentID, projectID) are documented in the schema with patterns and descriptions, so schema coverage is 100% and the baseline is 3. The description adds nothing parameter-specific – it never mentions ids, scoping, or which org/project must own the agent – so it neither compensates nor is required to.

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 verb+resource (archive the agent) and then enumerates the exact scope of that verb: runtime cancellation, queued inputs, processes, pending wakeups, machine bindings, pooled machines, cron triggers. It implicitly separates itself from siblings like agents_cancel by making clear that this is a lifecycle retirement operation, not a runtime-only cancellation, and the retention semantics ('event history stay readable by id', 'leave listings') further pin down what 'archive' means here.

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?

Usage is implied rather than stated: an agent would read this as 'use me to retire an agent while keeping its history', and the no-op note gives a re-invocation condition. However, there is no explicit when-to-use/when-not-to-use guidance against the nearest alternative, agents_cancel, nor any statement of prerequisites or permissions.

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