Mark Bluesky conversations as read

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

Annotations already mark this as non-read-only and non-destructive; the description adds value by revealing that the reported unread count may differ from the requested scope ('Reports the unread count — or the number of conversations — Bluesky reads back, not the one requested'). This is a meaningful behavioral nuance beyond the structured annotations. The 'Free' note adds cost transparency. No contradictions with the annotations.

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 compact and front-loaded with the core action, then scope, then use case, then return-value caveat, then cost. It avoids redundancy with the schema, though the 'Free.' at the end is borderline extraneous and the phrasing 'Bluesky’s own mark-all' is a bit wordy. Overall, every sentence carries purpose, but a slight trim would improve it.

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 tool with 3 optional parameters, no output schema, and straightforward semantics, the description covers the essential behaviors: scope (single vs. all), optional narrowing via status, the return-value difference, and the use case. It doesn't explicitly state what happens when all parameters are omitted but implies 'mark everything.' The schema fills in the rest. No critical missing context for an agent to call it correctly.

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% and each parameter description in the schema is detailed (e.g., convoId's 'OMIT to mark EVERY conversation', status's enum with refusal behavior). The tool description reinforces the convoId omission and status narrowing but doesn't add new semantic info beyond the schema—e.g., it doesn't mention messageId at all. With high schema coverage, the baseline is 3, and the description only marginally enhances parameter meaning (like the explicit 'from list_bluesky_convos' source for convoId, which is already in the schema).

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 opens with a specific action and resource ('Clear the unread count on Bluesky DMs') and immediately clarifies the two scope modes (one conversation vs. whole account), which distinguishes it from sibling tools like list_bluesky_convos or read_bluesky_dm. The verb 'mark' and resource 'convo' are explicit, and the 'not the one requested' return-value note further pins down its behavior.

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 a concrete trigger ('Useful after triaging an inbox') and explains the benefit relative to list_bluesky_convos, explicitly stating that using this tool prevents the next list_bluesky_convos from surfacing the same threads. It doesn't list explicit alternatives to avoid (e.g., 'use read_bluesky_dm if you only want to read a single message'), but the context and the schema's convoId guidance (from list_bluesky_convos) provide enough usage direction. No exclusions stated, but the guidance is solid.

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