agent-plugins

Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques

14 skills · hooks

Blocks CRUD on the machine-local auto-memory directory and redirects the agent to the repo's ./memory/ folder so memory is git-tracked and shared across users

2 skills · 2 commands · hooks

Entrepreneurship toolkit adapted from gstack: idea validation, market research, strategy review, and structured decision pipelines. No code, just thinking.

4 skills

C# + SolidWorks API skill: documentation-first workflow, code-quality patterns, learnings, and a helper script to locate the SolidWorks Interop redistributable assemblies.

2 skills · 1 command

MCP server for grounded SolidWorks XMLDoc API search, catalog browsing, and complete type, enum, example, and guide retrieval.

1 skill · 1 MCP server

Hosted MCP skill with a same-session watcher for GitHub PR CI, reviews, feedback, and terminal state.

1 skill · 1 MCP server

Skill that writes accessibility-focused alt text for images about to be posted on social media (Twitter/X, Bluesky, Instagram, LinkedIn, Mastodon, Threads, Facebook).

1 skill

Skill for provisioning Cloudflare accounts through Wrangler's temporary-account flow, claiming them in a headed browser, and isolating named Wrangler auth profiles.

1 skill

PreToolUse hook that injects a visible output separator between Bash commands joined by `&&` or `;` so per-command output is easy to read

hooks

Skill that transforms terse bug reports into well-structured GitHub issues via gh CLI

1 skill

Stop hook that detects hedging language suggesting shortcuts or deferred work and prompts the agent to surface assumptions

hooks

Blocks the WebFetch tool and redirects to Firecrawl + Browserbase MCP tools

hooks

Skill that establishes a 1Password CLI (op) session interactively via tmux when service-account auth fails

1 skill

A grounded writing-style skill for Pedro's X, Threads, Mastodon, and Bluesky posts

1 skill

PreToolUse hook that auto-injects --headed into `playwright-cli open` invocations in Bash/PowerShell commands and recommends a standard viewport size

hooks

Skill that fetches active (unresolved) PR comments from GitHub and formats them for LLM review and reply

1 skill

Stop hook that detects when the agent dismisses issues as unrelated or pre-existing and asks for evidence on each dismissal

hooks

PreToolUse hook that auto-fixes Windows+bash path pitfalls (backslash paths, /dev/stdin) before execution

hooks

Skill (/reset) that tears down harness agent state, protects normal worktrees, resets the current worktree to origin/main, and in --force mode removes linked worktrees before reclaiming main in the primary worktree.

1 skill

DeepSeek Harness plugin that adds per-tool and per-skill token estimates to the composer context usage menu.

OMP extension that stores each distinct effective system prompt and provider tool context as hidden session metadata for transcript indexing.