Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Agent Skill Publish Gate

    1

    Pre-flight audit for SKILL.md files to prevent marketplace rejections and optimize listing conversion.

    $14.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Agent Skill Publish Gate

    Agent Skill Publish Gate

    Example session with this skill installed

    Audit my SKILL.md. It fetches token prices from api.coingecko.com and saves them to a csv. I haven't declared any permissions yet and want to list it for $45.

    • Read your context and instructions
    • Compiled the agent publish gate

    I have completed the pre-flight audit of your SKILL.md file against the 7-gate marketplace standard.

    Pre-Flight Audit Report

    GateStatusFindings
    Gate 1: PermissionsFAILFound 'fetch' and 'save' keywords; network and file_write tools are not declared.
    Gate 2: HostsFAILapi.coingecko.com is used at runtime but missing from the host declaration list.
    Gate 3: Trigger DescriptionPASSDescription follows the "Use when... what it does" pattern effectively.
    Gate 4: FrontmatterPASSValid kebab-case naming and clean key-value pairs detected.
    Gate 5: Body StructureWARNMissing 'Example Output' section; this often leads to low-quality flags.
    Gate 6: SecretsPASSNo hardcoded API keys or sensitive placeholders detected.
    Gate 7: PricingWARN$45 is high for data fetchers; Agensi data suggests $14–29 for this category.

    Fix Checklist (Prioritized):

    1. Declare Tools: Add network and file_write to your marketplace submission form or permissions manifest.
    2. Add Hosts: Include api.coingecko.com in the declared runtime hosts list.
    3. Add Example: Insert an ## Example section showing a sample CSV output to improve quality score.
    4. Adjust Price: Consider lowering to $29.00 to align with the proven install range for dev tools.

    Next steps

    • Update the SKILL.md body to include the missing Example section.
    • Sync your marketplace submission metadata with the declared tools and hosts identified above.
    • Re-run this audit once the permissions block is updated.
    I have completed the pre-flight audit of your SKILL.md

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Struggling with marketplace rejections or low conversion on your agent skills? Most submissions fail not because the code is bad, but because the metadata and permission declarations are inconsistent. This skill acts as a pre-flight audit for SKILL.md files to ensure they meet the specific requirements of Agensi, ClawHub, and SkillHub before you hit submit.

    What it does

    • Permission matching scans your skill body for keywords like "fetch" or "write" to ensure network, file_write, and terminal tools are correctly declared.
    • Host validation extracts runtime domains to verify they are listed in your host declarations, preventing common "undeclared network access" rejections.
    • Trigger optimization audits your frontmatter description to ensure it follows the "Use when... what it does" pattern required for agent discovery.
    • Security scanning greps for hardcoded secrets, tokens, or undeclared destructive side effects that trigger immediate safety flags.
    • Pricing analysis compares your proposed price against Agensi market data to ensure you are within proven install-conversion ranges.

    How it works

    1. Provide the file by pointing the agent to your SKILL.md or the contents of your submission package.
    2. Execute the audit to run the 7-gate check covering frontmatter, structure, permissions, and hosts.
    3. Review the scoreboard which outputs a PASS/FAIL/WARN table across all submission categories.
    4. Implement fixes using the prioritized checklist generated at the end of the report.

    Frameworks & tools

    This skill is optimized for SKILL.md specifications used by Agensi, ClawHub, and SkillHub. It supports auditing skills built for Claude Code, Cursor, and MCP servers.

    Why this beats prompting it yourself

    Generic prompts often miss the specific permission-to-keyword mapping that marketplace reviewers use. This skill uses a specialized lookup table and Agensi pricing data to catch the exact technicalities that lead to manual rejection.

    Use cases

    • Auditing a new skill before the initial marketplace submission.
    • Debugging a "declared permissions mismatch" rejection from a marketplace moderator.
    • Optimizing an existing listing that has high views but low install conversions.
    • Packaging a local CLI tool or MCP server for commercial sale.

    Known limitations

    This is a static document audit, not a runtime sandbox test. It cannot detect hosts accessed via dynamic redirects or obfuscated code execution at runtime.

    Static audit only; does not execute code. Cannot detect hosts behind dynamic redirects or obfuscated strings. Requires SKILL.md format.

    How to use

    1. 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. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    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 1 month ago

    Permissions

    Network Access
    Read Files
    Write Files

    Static document audit: reads SKILL.md text (Read Files) and writes the audit report (Write Files). Network is declared defensively — the audit checklist references fetch/HTTP/RPC only as detection keywords it scans for in the document under review; the skill itself makes no runtime network calls.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    This skill automates the pre-submission review process, identifying missing permission declarations, security flags, and host mismatches that lead to immediate manual rejections by marketplace moderators.

    It is specifically designed for SKILL.md files used by Agensi, ClawHub, and SkillHub, and supports auditing skills intended for Claude Code, Cursor, and MCP servers.

    While it captures 95% of common metadata errors, it is a static document audit and cannot detect hidden network requests made through obfuscated code or dynamic runtime redirects.

    Yes, the audit includes a pricing analysis that compares your proposed cost against real-time market data to ensure your listing is priced for maximum conversion.

    No setup is required; simply provide the contents of your SKILL.md file or your submission package to the agent to start the 7-gate audit process.

    You receive the full 'agent skill publish gate' logic, including the specialized lookup tables for permission-to-keyword mapping and the prioritized fix checklist generator.

    New in Prompt & Skill Engineering