share_intent

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

With zero annotations provided, the description carries the full burden — and it excels. It discloses that the operation is 'ADVISORY and a CLAIM, not a lock', idempotency (one live intent, calling again replaces it), expiration and session-clearing semantics, the polling/hint-injection delivery guarantee, the config gating requirement, and the 'secret-scrubbed before storage' security behavior. This is exactly the behavioral disclosure an agent needs and far beyond what annotations would have given.

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?

Excellent front-loading: the first sentence is a model of purpose communication. But the body is dense and repetitive — the 'unverified claim' / 'claim' concept appears three times, 'workspace_sessions' twice, and jargon like 'plumb does not push' is unexplained, which will cost an agent comprehension time. The trailing parameter section duplicates the schema with minimal extra info. Well-organized but overstuffed for what could be 40% shorter.

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?

Despite the absence of annotations and output schema, an agent has everything required to call this correctly: preconditions, single-live-intent semantics, TTL defaults and overrides, session life-cycle behavior, peer-visible effects, and security handling. Given the tool's complexity — stateful, cross-agent, security-sensitive — there are no obvious informational gaps an agent would trip over.

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. The description adds a helpful concrete glob example ('internal/tools/ratelimit*') and connects path_globs to the 'drives peer write hints' behavior that the schema implies but doesn't state. However, the 'Parameters:' section in the description largely restates what the schema already documents, so the net new value is incremental rather than transformative. A 3 is the right call.

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?

Starts with a specific verb+resource: 'Broadcast what you are working on to other agents active on this workspace RIGHT NOW'. The 'steer around... instead of colliding' phrase gives the operational outcome, and the example ('refactoring the rate limiter') grounds it immediately. Even without reading siblings, an agent can tell this broadcasts intent for coordination rather than persisting or recalling data.

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?

Explicitly contrasts the tool with the sibling 'workspace_sessions' recent_writes, and states when the call is refused ('Requires [collab] intents = true'). It also clarifies the delivery model ('polling and hint injection only'). Minus one because it never names a sibling for the alternative case (e.g., persisting memory or checking messages), so an agent must infer when NOT to use it from the single workspace_sessions comparison.

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