Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    ChangeShield Pro | GitHub Release Gate

    1

    Verify exact-commit GitHub CI evidence and review nine release safety gates. Catch stale checks, missing approvals and risky deployments before shipping.

    $14.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    ChangeShield Pro | GitHub Release Gate

    ChangeShield Pro | GitHub Release Gate

    Example session with this skill installed

    Use ChangeShield Pro to evaluate the bundled synthetic release evidence in examples/blocked.json. Run the offline evaluator, explain the verdict and identify the blocking gate. Do not access GitHub, modify files, or deploy anything.

    • Read your context and instructions
    • Compiled the changeshield pro

    ChangeShield Release Review

    Verdict: NO-GO

    Repository: demo/example-service
    Target: Staging
    Evidence basis: Synthetic manifest, evaluated offline. No independent GitHub verification.

    Release gateResult
    Source integrityPASS
    Required CIFAIL
    Test suitePASS
    Security reviewPASS
    Secret screeningPASS
    Migration safetyNOT APPLICABLE
    Rollback planPASS
    ObservabilityPASS
    Human approvalPASS

    Why release is blocked

    The required CI gate has failed. Other passing gates cannot override this failure.

    Next safe step: Investigate the failed check, rerun the relevant tests on the exact release commit, and review fresh evidence before reconsidering approval.

    No deployment, repository changes, credential access, or network requests were performed.

    This is a synthetic example demonstrating the bundled offline evaluator.

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

    About this skill

    Stop green-check illusions before a release

    A CI workflow can be green while referring to an older commit. ChangeShield Pro helps developers and AI coding agents review release evidence for the exact commit they plan to ship.

    What it does

    • Verifies selected GitHub check results against the candidate commit and branch.
    • Detects missing, pending, failed, and mismatched CI evidence.
    • Reviews nine release safety gates, including tests, security, migrations, rollback, monitoring, and human approval.
    • Produces transparent NO-GO, CONDITIONAL GO, or provisional GO assessments.
    • Includes Python verification tools, sample evidence, security guidance, and 32 automated local tests.

    Designed for safe development workflows

    ChangeShield is read-only by design. It doesn't push code, merge pull requests, deploy applications, or modify production databases.

    Requirements and limitations

    The skill is designed for agents supporting the SKILL.md convention. Optional verification scripts require Python 3.10+ and permission to access GitHub's API.

    Successful CI checks alone do not authorize a production release. Final approval remains with the responsible human operator.

    Created by Mahmoud Karzoun

    Inspired by the MIT-licensed MK Nuclear Spacecraft project, with separately developed verification scripts, tests, and release-review workflows.

    Read-only. No deployment or git write capability. Requires Python 3 for helper scripts. SHA verification depends on user-provided or API-accessible evidence.

    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 27 days ago

    Requires

    GitHub
    Python

    Permissions

    Terminal / Shell
    Network Access
    Read Files
    Write Files
    Environment Variables

    Allowed Hosts

    api.github.com
    github.com

    Read-only release assessment. Terminal access is used to run bundled Python scripts. File reading is required for evidence manifests. Optional network access is limited to the GitHub API. An environment token is optional for public repositories and may be needed for private repositories. No deployment, repository modification, or credential storage is performed.

    Listed27 days ago

    What's inside

    Frequently Asked Questions

    No, this skill is strictly read-only and does not have the capability to merge PRs, trigger deployments, or modify your repository state. It acts as an independent auditor to provide a GO/NO-GO verdict based on existing evidence.

    The skill uses a dedicated Python helper to cross-reference every piece of evidence against a specific Git SHA. This prevents 'state drift' where a release is approved based on outdated CI results from a previous commit.

    Public GitHub repositories can usually be checked without a token, subject to API rate limits. Private repositories require appropriately scoped read-only access. Tokens are optional and must never be pasted into chat. Offline evidence evaluation requires no GitHub credentials or network access.

    This skill is compatible with any AI agent environment that supports Python 3 execution and external API calls, specifically those following the standard Agent Skills specification.

    You receive the complete evidence-gathering framework, including the Python-based CI observation helper, the deterministic policy evaluation engine, and the standardized release reporting template.

    While it is optimized for GitHub Actions, the skill can evaluate evidence from any CI tool that reports back to the GitHub Checks API or provides accessible build manifests.

    New in DevOps & Cloud