Publishes a package to its registry, returns an approval job, skips existing versions, and pauses publishing for 24 hours after an E429 rate-limit response.
Publish to the registry; returns a job (npm approval opens in a terminal). Skips existing versions, and refuses for 24 h after an E429.
| Name | Required | Description | Default |
|---|---|---|---|
| dir | Yes | ||
| force | No | Ignore the E429 wait; only if the user says the block was lifted | |
| dryRun | No | ||
| openBrowser | No |
Changes observed during successful MCP inspections.
v0.4.0Input schema / $schemaRemoved value: -"http://json-schema.org/draft-07/schema#"Input schema / properties / dir / descriptionRemoved value: -"Project folder (absolute path, or relative to the server's working directory)."Input schema / properties / force / descriptionPrevious value: -"Publish even within 24 hours of an E429. Only when the user confirms the block was lifted."New value: +"Ignore the E429 wait; only if the user says the block was lifted"Input schema / properties / openBrowser / descriptionRemoved value: -"Open approval pages in the default browser."v0.3.0Input schema / properties / forceAdded value: +{
+ "default": false,
+ "description": "Publish even within 24 hours of an E429. Only when the user confirms the block was lifted.",
+ "type": "boolean"
+}v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The annotations only declare destructiveHint=false, so the description carries most of the burden and does add real value: publishing returns a job, npm approval opens in a terminal, existing versions are skipped, and the tool self-blocks for 24h after an E429. It stops short of saying whether the publish is irreversible or what the job object contains.
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?
Dense and front-loaded: the core action and return value come first, caveats second. The telegraphic style and undefined 'E429' cost a little clarity but nothing is padded.
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 4-parameter mutation with no output schema, the description should cover auth prerequisites, what 'dryRun' and 'openBrowser' do, and how the returned job relates to job_status. It covers the E429/skip semantics but leaves those gaps.
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 only 25% — only 'force' has a schema description. The description explains the E429 behavior that makes 'force' meaningful, but says nothing about 'dir', 'dryRun', or 'openBrowser', so three of four parameters are undocumented in both places.
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?
States a specific verb and resource ('Publish to the registry') plus the return shape ('returns a job'), which is enough to distinguish it from siblings like preflight, check_names, or tag_and_release. It does not explicitly name which sibling to use first, but the action is unambiguous.
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 only implied: the note about skipping existing versions and refusing for 24h after an E429 hints at preconditions, but there is no explicit 'use this when X, use preflight/check_names first otherwise' guidance. Prerequisites such as being logged in (a login sibling exists) are unstated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.