general-plugin

Plugin general-plugin from marketplace vibe-coding in AndrewChang-cpu/vibe-coding-configuration. Commands read from the repo's manifest — 1 month ago.

Skills (23)

• vibe:add-tests

  Retrofits tests onto completed work. Classifies changed files as TDD (unit), E2E, or Skip, gets user approval, then generates tests following RED-GREEN conventions. See README for the distinction from vibe:work's built-in TDD.

• context-optimization

  This skill should be used when the user asks to "optimize context", "reduce token costs", "improve context efficiency", "implement KV-cache optimization", "partition context", or mentions context limits, observation masking, context budgeting, or extending effective context capacity.

• vibe:debug

  Systematic bug investigation. Collects symptoms from the user, delegates investigation to the debugger agent, presents root cause and proposed fix.

• doc-coauthoring

  Guide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc

• filesystem-context

  This skill should be used when the user asks to "offload context to files", "implement dynamic context discovery", "use filesystem for agent memory", "reduce context window bloat", or mentions file-based context management, tool output persistence, agent scratch pads, or just-in-time context loading

• frontend-design

  Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML/CSS layouts, or when styling/beautify

• vibe:plan

  Interactive planning — grills the user to produce a complete plan before any implementation. Works for greenfield and brownfield projects.

• python-patterns

  Living Python patterns and project standards — Protocol, dataclasses, context managers, async/await, type hints, dependency injection, package organization. Consulted automatically by vibe:work for Python files. Update this file as codebase standards evolve.

• python-testing

  Living Python testing patterns — pytest fixtures, parametrize, mocking, coverage, async testing, conftest structure. Consulted automatically by test-driven-development and add-tests skills for Python projects. Update as codebase-specific conventions emerge.

• request-recap

  Background hook — not a slash command. Fires automatically at session end to ask for a concise recap of the most recent response.

• vibe:review-fix-loop

  Codex-backed headless review-fix loop. Launches fresh codex exec phases for review, taskify, and work until clean, blocked, or user input is needed.

• vibe:review-fix-review

  Review phase of the review-fix loop. Runs vibe:review and normalizes findings into the review-fix ledger.

• vibe:review-fix-taskify

  Converts confirmed review-fix loop ledger tasks into executable .plan/TASKS.md entries for vibe:work. Codex headless wrapper; does not run implementation.

• vibe:review-fix-work

  Runs vibe:work for review-fix tasks until the current batch is complete, then syncs .plan/TASKS.md status back to the review-fix ledger.

• vibe:review

  Adversarial code review of current changes. Defaults to deep review. Spawns code-reviewer agent then applies pre-commit checklist as a second pass. Replaces vibe:audit.

• vibe:ship

  Prepares the current branch for a PR and pushes a draft PR. Generates PR title and body from .plan/PLAN.md. Warns if .plan/ transient files are tracked.

• vibe:tasks

  Decomposes .plan/ documents into a file-level task breakdown with dependency tags and an ASCII WBS graph. Run after /vibe:plan.

• test-driven-development

  Use before changing reusable application or library behavior in a real codebase with a test harness. Do not use for content edits, Markdown/link rewrites, metadata cleanup, file moves, generated output, configuration-only changes, or one-off data repair.

• vibe:verify

  Conversational UAT. Walks through each DoD criterion from the plan, asks the user if reality matches, records pass/fail in .plan/UAT.md, and summarizes results. Run after vibe:work.

• vibe:work-goal

  Runs vibe:work under /goal until every task in .plan/TASKS.md is done and every Definition of Done criterion in .plan/PLAN.md passes. Run after vibe:tasks.

• vibe:work-loop

  Schedules /vibe:work to run at a recurring cadence across sessions — survives usage limit resets. Run after /vibe:tasks when you want cross-session task execution.

• vibe:work-ralph

  Runs /vibe:work in a ralph-wiggum loop — one wave per iteration, sequential and inspectable. Requires the ralph-wiggum plugin. Run after /vibe:tasks.

• vibe:work

  Executes .plan/TASKS.md — dispatches implementer subagents for pending tasks, runs a single reviewer per task, updates statuses, and verifies DoD when complete. Run after /vibe:tasks.

Subagents (7)

• build-error-resolver

  Build and type error resolution specialist. Fixes errors with MINIMAL diffs only — no architectural edits, no refactors, no improvements. Goal is green build only. Invoke manually when a build or type-check fails.

• code-reviewer

  Adversarial code reviewer. Assumes defects exist. Accepts depth (quick/standard/deep) and file list via prompt. Outputs BLOCKER → WARNING → INFO findings to terminal. No files written.

• debugger

  Investigates bugs using the scientific method. User is the reporter (symptoms/expected/actual). Claude is the investigator (finds cause). Never asks the user what is causing the bug. Forms hypotheses, tests them, proposes fix.

• doc-updater

  Documentation sync specialist. Given a list of changed files, checks whether READMEs, docstrings, or other docs need updating to reflect the changes. Spawned automatically by vibe:work after all tasks complete.

• python-reviewer

  Python code review specialist. Runs ruff/mypy/black/bandit automatically, then reviews for CRITICAL/HIGH/MEDIUM issues with severity-tiered output. Blocks on CRITICAL or HIGH. Spawned automatically by vibe:review when Python files are in the diff.

• researcher

  Research agent. Investigates technical questions before planning. Uses Context7 → WebFetch → WebSearch priority. Tags every claim with provenance. Checks package legitimacy. Writes .plan/RESEARCH.md.

• security-reviewer

  Security vulnerability specialist. OWASP Top 10, secrets detection, injection, auth, XSS, dependency CVEs. Spawned automatically by vibe:review on every review pass.

MCP servers (6)

• github

  runs: npx -y @modelcontextprotocol/server-github

• postgres

  runs: npx -y @modelcontextprotocol/server-postgres postgresql://localhost:5432/postgres

• sequential-thinking

  runs: npx -y @modelcontextprotocol/server-sequential-thinking

• puppeteer

  runs: npx -y @modelcontextprotocol/server-puppeteer

• context7

  runs: npx -y @upstash/context7-mcp --api-key ${CONTEXT7_API_KEY}

• tavily

  runs: npx -y tavily-mcp

Hooks (3)

• UserPromptSubmit

  claude plugin update general-plugin@vibe-coding

• PostToolUse

  bash ${CLAUDE_PLUGIN_ROOT}/hooks/tdd-reminder.sh

• Stop

  bash ${CLAUDE_PLUGIN_ROOT}/hooks/request-recap.sh

Similar plugins