subscribe

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

With no annotations, the description carries full burden. It explicitly discloses that this is a 'Preview' and that persistent subscriptions won't come until 'Wave 3.' It clearly states the current behavior: returns current state plus advisory polling instructions, and gives concrete URLs (RSS at /changelog/feed.xml or polling /data/attestations/{slug}.json). This is excellent transparency—the agent knows it's not a real subscription and what to actually expect.

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 only two sentences but packs important info: preview status, current behavior, specific endpoints, and usage intent. It is front-loaded with the key caveat, then gives actionable details. Every phrase adds value without redundancy.

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?

This is a somewhat unusual tool (a placeholder for future functionality). The description covers what it returns, how to act on it, and the limitation. It lacks detail about the exact shape of the 'current state' but given there's no output schema, it's still fairly complete. The preview nature is fully explained, and the provided URLs are enough for an agent to proceed.

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 baseline is 3. The description adds value by explaining the 'channel' parameter in terms of the actual polling options (RSS feed and JSON polling URL) and by noting that 'webhook is advisory only until Wave 3 ships.' This goes beyond the schema, which only lists the enum values, and helps the agent understand what each channel will return. Thus a 4 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 clearly states what the tool does: 'returns the agent's current state plus advisory polling instructions' and indicates its purpose: 'register interest in being notified when a slug's verdict changes.' It is specific about the verb (returns, register) and resource (state, notifications), but the name 'subscribe' is somewhat misleading because it actually doesn't create a subscription yet. It does distinguish from siblings by focusing on interest registration rather than verification or scoring.

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?

The description says exactly when to use it: 'Use this to register interest in being notified when a slug's verdict changes.' This is clear context. It also mentions the preview limitation and tells users what they'll get instead of a persistent subscription. However, it doesn't provide explicit alternatives or when-not-to-use guidance, which would have pushed it to a 5.

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