More screenshots
Works with the AI Agents you already use
📦 npm Publish Payload Preflight
Review an npm pack file list for missing declared files, credential-like filenames and payload changes before publishing.
$5
📦 npm Publish Payload Preflight
Example session with this skill installed
{"package":{"name":"tiny-kit","version":"1.0.0"},"pack_files":["package.json","dist/index.js"]}
- Read your context and instructions
- Compiled the npm publish payload
- Generated the data export
Representative synthetic run, package v1.0.1
{"profile":"npm-publish-payload-preflight","gate":"REVIEW","recordsReviewed":2,"findingCount":1,"counts":{"warn":1}}
RUNTIME_INTENT_REVIEW: No root runtime entry is established. Build folders and type declarations alone do not prove loadability; confirm whether this is a runtime, types-only or data package.
The supplied manifest includes package.json and dist/index.js, but it does not declare a main, exports or bin entry, and no conventional root index file is listed. This is a review finding, not proof that the package is broken.
The run writes findings.csv, review-manifest.csv, result.json and report.md. Inspect the actual package metadata and test the intended import or usage before deciding whether to publish. The checker did not run npm, read source contents for secrets, contact a registry or publish anything. Exit 0 means the check completed, not that publishing is approved.
findings.csv
CSV · data export
Example file from a real run - the skill writes it into your workspace.
Connects securely to your tools. The creator never sees your data.
About this skill
npm Publish Payload Preflight turns owner-supplied local evidence into a deterministic preflight. A package can publish successfully while omitting built runtime files, leaking local configuration, or silently changing its shipped surface. It writes exact findings, a review manifest, a machine-readable gate, and a handoff report. It never performs the external action and never turns a passing check into a certainty.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
Does not run npm, publish packages, inspect file contents for secrets or prove Node import behavior. Filename screening is bounded and can miss secrets under other names. Conditional, pattern, escaped and legacy fallback entry points require intended-runtime review. Types-only and data packages require intent review rather than being assumed broken. READY means no findings within the documented checks, not approval to publish.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 25 days ago
- Software
Python 3.10+
The agent must be able to run Python. Supply your npm pack file list and package metadata.
Permissions
File Scopes
The checker is deterministic and self-contained.