Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Codex Project Readiness Gate

    1

    Check repository roots, instruction chains, and verification readiness before material Codex work.

    $9

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Codex Project Readiness Gate

    Codex Project Readiness Gate

    Example session with this skill installed

    Repository root: /work/acme-api
    Working directory: /work/acme-api/services/billing
    Declared validation command: pytest -q

    • Read your context and instructions
    • Compiled the codex project readiness

    Decision: needs_changes
    High: nested instruction chain exceeds the configured 32 KiB budget.
    Medium: declared verification command is not supported by observed repository markers.
    Unknown: runtime Codex behavior.
    Human verification required: review findings and any proposed patch before use.

    The actual result also includes deterministic machine-readable findings and a minimal proposed patch that is never applied automatically.

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

    About this skill

    Check a repository before the first material Codex task—or after its instructions, working root, or verification steps drift. This local, read-only skill traces the applicable AGENTS.md/override chain, checks the 32 KiB instruction budget, flags unsafe or unverifiable guidance, and returns a human report, deterministic JSON findings, and a minimal proposed patch.

    It does not run repository commands, read secret files, change the repository, use a network service, or guarantee what Codex loaded or will obey.

    Support covers installation, documented use, and package defects. It excludes repository consulting and assurances about Codex behavior. Buyers should verify the exact package hash before installation.

    Python 3.10+ is required. Findings about contradictions, staleness, roots, and command support use documented heuristics and require human review. The gate reads only applicable instruction files, explicitly selected non-sensitive guidance, and bounded path metadata—not general source code. It never executes validation commands, so it cannot prove a build or test passes. It cannot prove what Codex loaded or will obey. Metadata inventory stops after 10,000 entries; each inspected text file is limited to 1 MiB.

    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 3 months ago

    Permissions

    Terminal / Shell
    Read Files

    File Scopes

    **/AGENTS.md
    **/AGENTS.override.md

    Reads bounded path metadata under the buyer-supplied repository root and only the applicable instruction chain plus explicitly selected non-sensitive guidance. Never writes repository files.

    Listed3 months ago
    Updated3 months ago

    What's inside

    Frequently Asked Questions

    It acts as a pre-flight check to ensure your AGENTS.md files and instruction chains are within the 32 KiB budget and validly structured before you begin expensive or complex Codex tasks.

    No, this skill is strictly local and read-only; it analyzes your project structure and instructions without executing code, accessing your network, or modifying any repository files.

    You will receive the local skill package, a human-readable readiness report, deterministic JSON findings for CI/CD integration, and a suggested patch for any detected instruction errors.

    While optimized for Codex-based workflows and AGENTS.md standards, it can be used with any local agent environment that follows hierarchical instruction chains and needs to monitor context window budgets.

    The skill identifies if instructions are 'unverifiable' or 'unsafe,' such as guidance that is too vague for the agent to confirm or prompts that might lead to non-deterministic outcomes.