By pedropaulovc
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements and design before implementation.
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Use when you have a written implementation plan to execute in a separate session with review checkpoints
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
A Claude Code plugin marketplace of hooks, skills, and commands focused on engineering rigor, GitHub workflow, Windows quirks, and one very specific CAD niche.
/plugin marketplace add pedropaulovc/agent-plugins
/plugin install <plugin-name>@agent-plugins
Rust Stop hook that detects hedging language in the current turn, blocks the stop, and prompts Claude to report each assumption explicitly so you can make the judgement call.
Rust Stop hook that detects when Claude dismisses findings as unrelated or pre-existing, blocks the stop, and asks for evidence on each dismissal.
The only Claude Code skill targeting the SolidWorks .NET COM API. Anti-hallucination guardrails tuned for the low-level surface: many SolidWorks methods take 10–30 positional bool/int/double parameters where a flipped bool silently changes behaviour, so the skill forces named arguments and grounded references to the offline API docs over guesswork. Also: documentation-first workflow, COM-interop code-quality patterns, real-bug learnings (FeatureCut4 returning null, extrusion failures, faulty-geometry detection), and a /download-solidworks-docs command that pulls the offline API doc bundle into the skill folder.
Skill for writing social-media alt text that a screen reader user would actually want to hear. Pushes back on the default AI failure mode — exhaustive, forensic descriptions that read like a deposition — and instead frames every image around "what would the post lose if the image disappeared?". Bakes in platform-specific length budgets (Bluesky 2k, Mastodon 1.5k, X 1k, LinkedIn 120), forces transcription of any visible text (memes, tweet screenshots, chart labels), leads with the takeaway for charts, and avoids the common pitfalls of assigning identity from appearance and editorializing ("beautiful", "stunning").
Broadly useful regardless of what you're working on.
| Plugin | Type | What it does |
|---|---|---|
| superpowers | Skills | Core skills library — TDD, debugging, collaboration patterns (vendored from obra/superpowers) |
| windows-bash-guard | Hook | Auto-fixes Windows+bash path pitfalls (backslash paths, /dev/stdin) before execution |
| memory-to-repo | Hook | Blocks CRUD on the machine-local auto-memory dir and redirects to the repo's ./memory/ folder so memory is git-tracked and shareable |
Tuned to my own setup, tooling, workflow preferences, or niche — unlikely to appeal to a broad audience.
| Plugin | Type | What it does |
|---|---|---|
| mediocrity-detector | Hook | Detects hedging on Stop and pushes back |
| unrelated-issue-detector | Hook | Demands evidence for each "unrelated/pre-existing" dismissal |
| gh-issue | Skill | Turns terse bug reports into well-structured GitHub issues via gh |
| pr-comments | Skill | Fetches unresolved PR comments formatted for LLM review and reply |
| command-chain-separator | Hook | Injects a visible separator between Bash commands joined by && or ; so per-command output is easy to read |
| developing-solidworks | Skill + Command | C#/SolidWorks .NET COM API workflow with anti-hallucination guardrails |
| gstack-entrepreneur | Skills | Entrepreneurship subset of gstack: idea validation, market research, strategy (no code) |
| no-fetch | Hook | Blocks WebFetch and redirects to my Firecrawl + Browserbase MCPs |
| worktree-reset | Skill | /m — resets the current worktree to origin/main and reinstalls deps |
| playwright-cli-headed | Hook | Auto-injects --headed into playwright-cli open invocations and recommends a standard viewport |
| alt-text | Skill | Writes accessibility-focused alt text for images about to be posted on social media |
MIT
npx claudepluginhub pedropaulovc/agent-pluginsStop hook that detects when Claude dismisses issues as unrelated or pre-existing and asks for evidence on each dismissal
Stop hook that detects hedging language suggesting shortcuts or deferred work and prompts Claude to surface assumptions
Skill that transforms terse bug reports into well-structured GitHub issues via gh CLI
Blocks the WebFetch tool and redirects to Firecrawl + Browserbase MCP tools
PreToolUse hook that auto-fixes Windows+bash path pitfalls (backslash paths, /dev/stdin) before execution
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Superpowers Plus core skills library for Claude Code: planning, execution routing, TDD, debugging, and collaboration workflows
Harness-native ECC plugin for engineering teams - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, MCP conventions, and operator workflows for Claude Code plus adjacent agent harnesses
Claude harness - A harness for solo developers (Vibecoders) to handle full-cycle contract development.
Intelligent prompt optimization: injects the right context at the right moment so Claude lands a better first output. Clarifies vague prompts with research-based questions, plus targeted nudges for approach selection, plan readability, workflow routing, background execution, subagent routing, output readability, user-decision questions, and plan-mode assessment
Plugin-safe Claude Code distribution of Antigravity Awesome Skills with 1,640 supported skills.