sammcj/agentic-coding
Overview
This skill helps you discover, inspect, and evaluate open agent skills so you can decide whether to install them. It searches the skills registry, reviews source materials, and summarizes safety and functionality concerns. The output is an actionable briefing you can use before any installation.
How this skill works
I search the open skills ecosystem (using npx skills find and skills.sh) for candidates matching your query, then inspect each candidate's documentation and repository metadata. I flag risky patterns, permission scopes, and behavioural concerns, and I can optionally run or recommend an automated security scan before you ever install anything. I always present findings and installation commands for you to run — I do not install or bypass confirmations.
When to use it
- You ask “is there a skill that can X” or “find a skill for X”.
- You want to extend agent capabilities for a specialised task (e.g., CI, infra, testing).
- You’re unsure whether a task needs a dedicated skill or can be handled directly.
- You want a pre-install security and scope review before adding a dependency.
- You need help locating tools, templates, or workflows in the agent skills ecosystem.
Best practices
- Clarify domain and specific task before searching to use precise keywords.
- Treat skills as code: verify source repo, authorship, and license before installing.
- Never install without explicit user confirmation and never use -y/--yes flags.
- Watch for red flags: obfuscated code, excessive permissions, hidden instructions.
- Offer an optional security scan and inform the user if content will be sent to a third-party API.
Example use cases
- Find a skill that automates React integration test generation and report scope/risks.
- Search for a CI/CD helper skill, review whether it executes shell commands or network calls.
- Locate a knowledge-only skill for best-practice guidelines and confirm it has no execution surface.
- Investigate a skill a teammate recommended and run a Snyk Agent scan before install.
- When nothing matches, report no skills found and help complete the task directly.
FAQ
No. I present the install command and safety findings; you must run npx skills add yourself or explicitly ask me to run it with confirmation.
Can you run an automated security scan?
Yes — I can run the bundled scan script, but I will tell you first that the skill content is sent to the scanner API and require your permission.
22 skills
This skill helps you discover and evaluate open agent skills, then present findings to decide on installation.
This skill helps you create and refine diagrams of cloud infrastructure using Python diagrams library, producing ready-to-share images.
This skill helps you author and optimize skills efficiently by guiding structure, loading levels, and ensuring reliable activation for Claude.
This skill helps you create, review, and improve data visualisations using SWD principles for clearer charts, narratives, and decision-ready reports.
This skill helps you discover and validate configuration schemas from schemastore.org to ensure correct structure and supported options.
This skill creates structured development plans with phased tasks, requirements, and QA checklists to guide project execution.
This skill helps you set up and troubleshoot Home Assistant automations and dashboards by generating YAML configurations and inspecting entities.
This skill guides you to apply Mantel brand styles to diagrams and frontend components for consistent visuals.
This skill helps you create and update mermaid diagrams following best practices, including color themes, labeling, and formatting guidelines.
This skill reverse-engineers a design system from code and screenshots to produce a developer-ready design reference for components.
This skill defers task execution by scheduling a background timer and confirming details before running the requested action.
This skill guides you to create MCP Apps by registering tools and resources, wiring UI to hosts, and choosing framework templates.
This skill guides post-development code reviews by orchestrating parallel sub-agents to surface findings and enforce fixes.
This skill helps you train and export Piper TTS voices for offline deployment, covering data prep, validation, fine-tuning, and ONNX export.
This skill helps create and maintain CLAUDE.md project memories by capturing non-obvious conventions, patterns, and gotchas beyond code inspection.
This skill guides you in building and deploying AI agents with AWS Strands Agents SDK and Bedrock AgentCore, including architecture, observability, and
This skill helps you analyse written content using rigorous logical reasoning, producing concise, actionable critique that preserves domain insights.
This skill extracts key insights, actionable takeaways, and notable quotes from YouTube transcripts, blogs, and texts to inform decisions quickly.
This skill helps you evaluate LLM applications using Deepeval's pytest-based metrics and tracing to ensure quality and safety.
This skill analyzes a team session export against Claude Code best practices and delivers a structured, evidence-based improvement report.
This skill coordinates a multi-agent ideation session to convert a seed concept into actionable briefs without generating ideas itself.
This skill helps you create and maintain CLAUDE.md project memory files with non-obvious conventions, patterns, and gotchas to guide AI coding agents.
