Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Done Means Verified: Report What Was Actually Checked

    1

    Use when an agent says a task is done. Turns existing logs and checks into a completion report: each claim, the check behind it, what failed, what never ran, and whether more testing is needed.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Done Means Verified: Report What Was Actually Checked

    Done Means Verified: Report What Was Actually Checked

    Example session with this skill installed

    Review the latest build logs for the authentication module and provide a verification claim. The goal was to verify the OAuth2 flow is functional in the staging environment.

    • Read your context and instructions
    • Compiled the done means verified

    I have analyzed the existing check records for the authentication module. The verification claim is outlined below based on the staging environment logs from branch feat/oauth-fix.

    Verification Claim: Authentication Module

    Original Goal: Verify OAuth2 flow functionality in staging.
    Conclusion: Partially Verified. The handshake and token exchange are confirmed, but the final callback redirection to the dashboard remains unverified.

    ClaimActual Check and ResultLayerContext
    Provider ConnectivityGET /auth/provider returned 200IntegrationStaging, v2.4.1
    Token ExchangePOST /auth/token returned valid JWTLocal SampleStaging, v2.4.1
    Dashboard RedirectCheck not found in log (Timeout)Business End StateStaging, v2.4.1

    Gaps:

    Unknown: The final redirect to /dashboard did not appear in the logs. No error was recorded, but the session termination log is missing.

    • Not Executed: Post-login profile hydration check was skipped due to the redirect timeout.

    Necessary next step:
    Query the staging ingress logs for 302 redirects matching the session ID auth_8829 to confirm dashboard handoff.

    Next steps

    1. Fetch ingress logs for the timestamp 2023-10-27T14:30:05Z.
    2. Inspect DashboardController for potential silent failures during session initiation.

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

    About this skill

    Software verification often suffers from overconfidence where a successful build is mistaken for a working feature. This skill enforces a strict verification layer that separates static passes, mock results, and actual environment execution to prevent false positives in reporting.

    What it does

    • Evidence mapping creates a structured table linking specific claims to their actual check results and execution layers.
    • Layered validation categorizes results by scope: static structure, local samples, component integration, or business end states.
    • Gap identification explicitly flags failed, not executed, and unknown states using original error logs rather than summaries.
    • Status attribution tracks versioning, branch, and environment context for every check to prevent stale data usage.
    • Next-step logic generates singular, actionable completion signals required to close specific verification gaps.

    How it works

    1. Analyze logs by reviewing existing check records and execution outputs without running new tests.
    2. Classify layers to determine if the evidence matches the structural, functional, or business requirements.
    3. Generate table mapping the original goal to conclusion states like met, partially verified, or unverified.
    4. Finalize status by identifying the specific action needed to reach 100% verification or issuing a stop command if complete.

    Frameworks & tools

    This skill is platform-agnostic and works with CI/CD logs, test runner outputs (Jest, PyTest, Cypress), and manual QA records. It integrates with tools that provide draft or submission capabilities for reporting.

    Why this beats prompting it yourself

    Standard prompts often hallucinate that a successful linting pass means the feature works. This skill uses a rigid logic gate that forbids inventing evidence and forces the agent to admit when a business action is unverified despite technical success.

    Use cases

    • Auditing deployment logs to ensure all integration points were actually hit.
    • Reviewing PR status to confirm if unit tests covered the specific business logic changed.
    • Generating status reports for stakeholders that distinguish between "code complete" and "verified in prod".

    Known limitations

    This skill does not execute new tests or code. It relies entirely on existing records and cannot verify side effects that were not logged or captured during the initial run.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Testing & Debugging.

    Does not execute new code or tests. Cannot verify side effects if logs are missing. Limited to analyzing existing check records and environment metadata.

    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 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 7 days ago

    Permissions

    No special permissions declared or detected

    Listed7 days ago

    What's inside

    Frequently Asked Questions

    This skill creates a rigorous audit trail that prevents "false positives" by distinguishing between technical success (like a passing build) and actual business logic verification. It ensures that stakeholders know exactly which layers of a feature were tested and where gaps remain.

    No, this skill is purely analytical. It processes existing logs and records from your test runners (like Jest or PyTest) to classify results and identify missing evidence without running any additional code.

    This skill is platform-agnostic and designed to work with any AI agent that has access to text-based logs or reports. It can be easily integrated into workflows involving CI/CD pipelines, PR review bots, or documentation agents.

    Your purchase includes the full logic for evidence mapping and layered validation, along with documentation on how to feed your specific log formats into the skill for optimal reporting.

    The skill identifies a "gap" whenever a business requirement lacks direct evidence from the appropriate execution layer, such as a feature marked 'complete' that only has unit test coverage but no integration logs. It flags these as 'unverified' rather than assuming success.

    New in Testing & Debugging