- Home
- Skills
- Agents & Orchestration
- agent-skills
Works with the AI Agents you already use
Agent Skills
Portable Agent Skills: project-incubation, python/typescript/nodejs/react code-review, ci-cd-plumber, pr-review, postgresql-review, ui-system-review, regulatory-compliance-applicability-scan.
Free
agent-skills
Example session with this skill installed
Run a full Python code review on the current repository and check if the new CI/CD workflow in .github/workflows/main.yml follows security best practices.
- Read your context and instructions
- Compiled the agent-
I have completed a comprehensive audit of the Python codebase and the CI/CD pipeline configuration.
Audit Results
1. Python Code Review Summary
| Domain | Score | Status |
|---|---|---|
| Security | 6/10 | Needs Attention |
| Performance | 8/10 | Pass |
| Concurrency | 5/10 | Fail |
| Observability | 9/10 | Pass |
2. CI/CD Pipeline Security Audit
# Findings for .github/workflows/main.yml
- severity: critical
issue: "Workflow uses broad GITHUB_TOKEN permissions."
remediation: "Define explicit 'contents: read' permissions at the job level."
- severity: important
issue: "Missing checksum verification for third-party actions."
remediation: "Pin actions to specific commit SHAs instead of tags."
3. Critical Code Findings
- File:
services/processor.py| Line: 42 - Issue: Potential race condition in
asyncblock; shared state updated without locks. - Domain: Concurrency & Async Correctness
Next steps
- Apply the suggested permission hardening to
main.ymlto secure the CI environment. - Implement a
threading.Lockorasyncio.Lockinprocessor.pyto address the concurrency failure. - Run
python-code-review diffafter these changes to verify the score improvement.
Connects securely to your tools. The creator never sees your data.
About this skill
Engineering teams often struggle with inconsistent delivery pipelines, security gaps in production code, and fragmented repository structures. This package provides a suite of 11 portable skills designed to standardize and audit the entire software lifecycle, from initial repository incubation to live database health checks. It automates high-intensity tasks like multi-domain code reviews for Python, TypeScript, Node.js, and React, alongside specialized audits for CI/CD pipelines, PostgreSQL instances, and UI system consistency.
What it does
- Project Incubation scaffolds new repositories with best-practice architectures and audits existing structures against defined baselines.
- Multi-Stack Code Review executes deep-dive audits across 11 domains including security, performance, and async correctness for Python, Node.js, and React.
- CI/CD Plumbing builds production-grade pipelines and audits existing YAML workflows for security, speed, and reproducibility.
- Live Database Audits connects to PostgreSQL instances via MCPg to analyze schema integrity, indexing, and workload health.
- Regulatory Scanning identifies applicable privacy and fintech regulations (GDPR, CCPA, HIPAA) based on codebase signals and primary legal citations.
How it works
- Discovery involves the skill scanning the repository for configuration files like
python-review-config.tomlordocs/ci-cd-baseline.mdto determine scope. - Analysis runs sequential or domain-specific checks using reference assets for standards compliance, security patterns, and architecture.
- Reporting generates scored Markdown reports with file-and-line findings, severity rankings, and suggested (but not applied) remediations.
Frameworks & tools
This skill works with GitHub Actions, GitLab CI, CircleCI, PostgreSQL (via MCPg), React 18/19, Next.js, Node.js, and Python. It supports cross-platform UI audits for Web, SwiftUI, and Jetpack Compose.
Why this beats prompting it yourself
Generic prompts often miss domain-specific edge cases like async race conditions in Node.js or design token drift in UI systems. These skills use structured reference libraries and sequential domain scoring to ensure zero-hallucination audits that follow professional engineering standards.
Use cases
- Hardening a GitHub Actions pipeline before a major production release.
- Performing a pre-merge security and performance audit on a large Python PR.
- Checking a new health-tech project for HIPAA and GDPR applicability gaps.
- Standardizing a design system across React web and SwiftUI mobile apps.
Known limitations
PostgreSQL reviews require the MCPg tool for full live database health analysis. UI audits are currently focused on code-level consistency and do not include subjective visual critiques or full WCAG certification.
PostgreSQL audits require MCPg for live data. UI reviews check code patterns, not visual aesthetics. Regulatory scans suggest citations but do not provide legal certification.
How to use
- 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
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 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
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 21 days ago
- Free with an Agensi account
Requires
Permissions
Allowed Hosts