Compliance Scanning on the System Running

    1

    Scan live production for compliance breaches: every rule carries its basis, severities stay meaningful, the alert load fits the team.

    $49

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Compliance Scanning on the System Running

    Example session with this skill installed

    Validate a new rule for a rental portfolio: a 'CRITICAL' severity rule for missing fire safety certificates (loi), citing 'Code de la construction L123-1', with 2% expected prevalence.

    • Read your context and instructions
    • Compiled the compliance scanning on

    The rule catalog validation for the rental portfolio has been completed. The new rule meets all architectural requirements for basis, severity budgeting, and remediation linkage.

    Rule Validation Result

    Rule CodeScopeSeverityBasisStatusNotes
    FIRE-CERT-01lotcritiqueloiPASSEDReference CCH L123-1 verified.
    BAIL-DPEbailcritiqueloiPASSEDStable at 5% prevalence.
    CLE-PROPREclientinfovaleurPASSEDCorrectly tagged as company value.

    Alert Load Simulation

    • Total Records: 10,160
    • Estimated Monthly Alerts: 203 (based on 2% prevalence for new rule)
    • Team Capacity Status: GREEN (Estimated 16.5 hours of clearance work / month)

    Next steps

    1. Update regles.json with the new FIRE-CERT-01 entry.
    2. Run python scripts/valider_regles.py regles.json --json in your CI pipeline to lock the configuration.
    3. Deploy the updated scanner to the staging environment to verify the read-only adapter performance.

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

    About this skill

    Alerts an operator clears, not a dashboard nobody opens.

    Scan a live production system for compliance breaches and turn them into alerts an operator will actually clear - a rule catalogue where every rule carries its basis in law, regulation, contract or company values, severities that stay meaningful, an alert load sized against the team, remediation links and an audit trail. Use when a business must continuously check its own records against the rules it is bound by, when a compliance scan produces more alerts than anyone reads, or when an inspection asks how you know you are compliant.

    Four failures sink these projects, and none is a bug: every rule is critical so nothing is; a rule has no stated basis so nobody can defend it; a company preference is presented as a legal obligation; the volume exceeds what the team can clear, so the oldest alerts are never read and the scan becomes decoration.

    The distinction that makes the product is the basis. A rule rests on law, on a regulation, on a contract, or on your own company values - and a company-value rule is legitimate but must never be displayed as a legal obligation. Once an operator discovers one alert was a preference wearing a law's clothes, they discount all of them.

    Domain-agnostic, and verified as such: the same rule format and the same code were run on a rental portfolio (leases, lots, clients, accounting entries - 226 alerts a month, 22.6 hours against 16 available, flagged as a designed-in backlog) and on a healthcare system (patient records, prescriptions, access logs - 190 alerts, 12.7 hours against 40, clean). Adapting it to your system means four things and nothing else: name your scopes, write a read-only adapter per scope, write the rule catalogue, give alerts a stable identity so a second scan updates instead of duplicating.

    Tested scripts: scripts/valider_regles.py simulates the load before you run anything and refuses a catalogue that cannot work - a rule with no reference, a company value marked critical, a critical alert with no link to the record to fix, a rule that can never fire, a rule firing on most of the portfolio, severity inflation above a quarter, volume beyond the team's hours, duplicate codes. Ten finding types, each verified.

    Includes an EXPERT.md reference (exact parameters, thresholds, failure diagnosis): the rule schema field by field, severity as a budget, the load formula with two worked domains, adapter properties, alert identity and lifecycle with expiring exceptions, the audit trail an inspection actually asks for, six metrics defined so they cannot flatter you, and an eight-row failure table.

    Acceptable use: scanning systems your organisation operates or is mandated to audit, against rules you can cite. The skill refuses presenting an internal preference as a legal obligation, using the scan to monitor individual employees rather than records, scoring people, widening scope beyond the mandate, and automatic remediation that changes production without a human decision.

    Not legal advice: whether a rule states the law correctly is a question for a professional in your jurisdiction. The script reads no production data - only your catalogue and your counts.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for LLM Ops.

    Read-only by design. No automatic remediation. Requires manual mapping of legal texts to rules. Does not verify the legal accuracy of provided references.

    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 27 days ago

    Software
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed27 days ago

    What's inside

    Frequently Asked Questions

    It solves the problem of 'alert fatigue' and lack of accountability by enforcing severity budgets and requiring every automated check to be linked to a specific legal or regulatory basis.

    The skill includes the core Python logic for the scanning engine, a JSON catalog template for rule authoring, and a validation script to simulate alert loads before you go live.

    No, this skill is designed to work with standard Python 3 and read-only database adapters, meaning it can integrate into your existing LLM-ops workflow without requiring third-party cloud dependencies.

    Absolutely; the system is domain-agnostic and allows you to define your own 'scopes' and 'adapters,' making it just as effective for healthcare data privacy as it is for real estate compliance.

    The scanner identifies and logs discrepancies with stable IDs for audit trails, but it does not modify your production data; remediation must be handled by your team or a separate automation.

    New in LLM Ops