Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    production readiness audit

    1

    A multi-lens production readiness auditor that generates evidence-backed go/no-go reports and local dashboards.

    Free

    1 useSecurity scanned

    production readiness audit

    Example session with this skill installed

    Run a production readiness audit on this repository. Use parallel mode and start the dashboard. Staging URL is https://staging.api.example.com.

    • Read your context and instructions
    • Compiled the production readiness audit

    I have initialized the production readiness audit for the current repository. The audit is running in

    parallel mode, and the local dashboard has been started to track progress in real-time.

    Audit Directory: .readiness-audit/
    Git Ref: main (clean)
    Dashboard URL: http://127.0.0.1:8080

    Initial Audit State

    StageStatusDetails
    0: PreflightCOMPLETEDEnvironment verified; dashboard server live.
    1: ContextIN_PROGRESSInferring RTO/RPO from existing README.md.
    2: EvidencePENDINGScheduled 92 deterministic control probes.
    3: LensesPENDING8 lenses (Security, Backend, etc.) queued.

    The runtime intake has been successfully recorded for https://staging.api.example.com using the readonly-auditor role. A reachability check confirmed the target is responsive. I am now proceeding to Stage 1 to define the architecture map and threat boundaries.

    Next steps

    1. Review context.md once Stage 1 completes to verify my assumptions on data criticality.
    2. Keep the dashboard open at http://127.0.0.1:8080 to monitor lens findings as they arrive.
    3. Add .readiness-audit/ to your .gitignore to keep the audit trail local.

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

    About this skill

    Production readiness is often a subjective "gut feeling" that leaves teams vulnerable to outages and exploits. This skill replaces guesswork with a formal eight-lens audit that evaluates whether a codebase is actually ready for real users, real attackers, and real load. It treats the absence of critical systems—like monitoring, rate limiting, or backups—as high-severity findings, ensuring you never ship a system that is fundamentally incomplete.

    What it does

    • Multi-perspective auditing scans the repo using eight specialized lenses including security, backend, devops, and database.
    • Evidence-based reporting requires every finding to cite specific code lines or ledger entries, preventing vague or hallucinated issues.
    • Automated dashboarding launches a local web interface to visualize the audit trail, findings, and final go/no-go verdict.
    • Absence detection uses a dedicated ledger to track what was searched for but not found, flagging missing infrastructure-as-code or controls.
    • Runtime validation performs read-only inspections of live staging environments to confirm environment-specific configurations.

    How it works

    1. Context Intake: Define the system criticality, RTO/RPO targets, and threat model to calibrate finding severity.
    2. Evidence Pass: Run deterministic probes to build a semantic map and inventory of existing controls and missing patterns.
    3. Lens Dispatch: Launch parallel specialist agents to evaluate the evidence and generate tagged findings (CONFIRMED, NOT FOUND, or UNVERIFIED).
    4. Validation & Reporting: Verify findings against formal rules and generate a CTO-ready Markdown report and JSON verdict.

    Frameworks & tools

    This skill utilizes a suite of internal Python scripts for state management (audit_state.py), evidence gathering (evidence_scan.py), and validation (validate_findings.py). It is designed to run in environments like Claude Code or Cursor with access to local files and a Python 3 runtime.

    Why this beats prompting it yourself

    Generic prompts often miss what is NOT there, which is where most production failures start. This skill enforces a strict evidence ledger and a multi-agent "lens" architecture that prevents a single agent from overlooking cross-disciplinary gaps. It provides a persistent, structured audit trail under .readiness-audit/ rather than a fleeting chat response.

    Use cases

    • Auditing a repository before a major production launch or migration.
    • Performing a technical due diligence review of a new codebase.
    • Identifying scalability bottlenecks and security gaps during a 10x load prep.
    • Providing an objective go/no-go report for engineering leadership.

    Known limitations

    This skill is strictly read-only and will not fix the issues it finds. It requires a local Python 3 environment to run the underlying orchestration and dashboard scripts.

    Read-only audit; no automated fixes. Requires local Python 3. Effectiveness on non-standard stacks depends on the map.md architecture assessment. Skips AI security if no model calls are found.

    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

    1 use

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 1 month ago

    • Free with an Agensi account

    Requires

    GitHub
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Browser
    Network Access
    Environment Variables

    Allowed Hosts

    raw.githubusercontent.com
    staging.example.com
    Listed1 month ago

    What's inside

    Frequently Asked Questions

    The audit focuses on eight critical lenses: Security, Backend Architecture, DevOps/Infrastructure, Database Integrity, Observability, Performance/Load, Reliability (RTO/RPO), and Compliance Readiness.

    Unlike generic chat prompts, this skill uses a dedicated 'absence detection' ledger to identify missing critical infrastructure, like rate limiting or backups, rather than just reviewing existing code.

    The skill is designed for terminal-based AI agents like Claude Code or Cursor that have access to a local Python 3 runtime and your project files.

    You receive the core orchestration logic, the eight specialized audit lenses, internal Python scripts for evidence gathering, and the local dashboard visualization tool.

    Yes, the auditor is strictly read-only; it scans your files and staging environments to identify risks but will never modify your source code or infrastructure.

    The skill produces a CTO-ready Markdown report, a structured JSON verdict for CI/CD integration, and a local web dashboard to visualize the audit trail and evidence.

    New in LLM Ops