Publish Skill

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

Annotations supply only readOnlyHint=false, so the description carries most of the weight and does add real behavior: the server performs the file read itself, and republishing the same name produces a new version rather than failing. That versioning semantics is the key non-obvious trait for a write tool. Still missing: what happens to the prior version's files, whether auth or space permissions are required, and what the call returns.

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?

Four short sentences, front-loaded with what the tool does and immediately followed by the operational instruction. Every sentence carries a distinct, actionable fact (what gets published, how to pass the folder, versioning, and the anti-pattern to avoid). No filler or restatement of the name/title.

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 mutation tool with six optional parameters, no output schema and a bare annotation set, the description covers the main risk areas: input mode selection, server-side reading, and version bumping. The remaining gap is the post-call picture — no mention of what is returned (version number, ID, URL) or what the agent should do next to install the published skill.

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 the schema already documents all six parameters, making 3 the baseline. The description adds modest beyond-schema meaning by framing `directory` as the preferred input that the server reads byte for byte and warning against mirroring it into `files`. It adds nothing about `name` defaulting, `note`, `spaceId`, or `baseVersion` beyond what the schema says.

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 states a specific verb and resource ('Publish a Claude Code skill folder (SKILL.md plus its references and scripts) to SystemDox') and names the downstream outcome ('installed on any computer'). An agent can tell this is an upload/publication operation rather than a read. It does not explicitly contrast with the sibling 'install_skill' or 'publish_setup_part', which is the only reason it falls short of a 5.

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 an explicit positive path ('Pass `directory`... the server reads the files itself, byte for byte') and an explicit negative rule ('Never retype the files into `files` when the folder is on this computer'), which resolves the directory-vs-files ambiguity an agent would otherwise hit. It also states that republishing a name creates a new version. It gives no guidance on when to prefer this over the sibling install_skill or publish_setup_part.

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