Take a Markdown implementation plan, run it through a parallel agent swarm with per-wave verification, and generate a bug-fix plan for re-runs; optional TDD red-green mode with per-task red/green gate evidence
plan-runner pipeline agent that reads all per-wave bug JSONs from a pipeline run, deduplicates and ranks bugs (P0-P3), produces a human-readable bugs.md summary, and generates a free-form Markdown fix-plan.md that is valid input for a re-run of /plan-runner:run.
plan-runner pipeline agent that reads a free-form Markdown implementation plan and returns a structured wave plan: file-disjoint task buckets of <=6 agents per wave, ordered as a DAG so each wave can execute in parallel without write conflicts.
plan-runner pipeline agent that implements a single task from a wave plan. Generic template -- the orchestrator parameterizes each invocation with the specific task title, excerpt, owned files, acceptance criteria, and Context7 flag.
plan-runner pipeline agent that writes ONLY the failing tests for one testable task from a wave plan. It never writes implementation -- a downstream impl agent makes the tests pass. Returns the test files and test IDs it added.
plan-runner pipeline agent that verifies ALL dev agents in a wave against their acceptance criteria. Reads each agent's owned files and dev return data; flags every gap as a structured bug entry in a single wave-level bug report.
Uses power tools
Uses Bash, Write, or Edit tools
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.
Five production-grade plugins for Claude Code: multi-agent code review, queryable architecture maps, parallel plan execution, spec consolidation, and vulnerability-aware dependency upgrades.
# One-time: add the marketplace
claude plugin marketplace add MisterVitoPro/qa-claude-market
# Then install whichever plugins you want (see per-plugin sections below)
claude plugin install qa-swarm@mistervitopro-plugin-marketplace
A swarm of specialist agents reviews your code, then fixes the bugs test-first.
Six Sonnet core reviewers (security, correctness, performance, architecture, data-flow, async) run in parallel, optionally joined by up to six Haiku specialists (config/env, supply chain, type safety, state mgmt, logging, backwards-compat). Findings are deduplicated, ranked P0–P3, and corroborated across agents. The implement skill picks up the report and fixes issues with a 3-agent TDD loop (failing test → minimal fix → verify).
claude plugin install qa-swarm@mistervitopro-plugin-marketplace
/qa-swarm:attack "audit the authentication flow for security and correctness"
/qa-swarm:implement
Stops Claude from re-exploring your repo every session.
Three analyst agents scan your codebase in parallel and a graph synthesizer annotates the key modules, producing a curated architecture index (directory map, key files, tech stack, patterns, dependencies, build commands) plus a semantic dependency graph. Graph queries (dependencies, dependents, blast radius, risk filters) run deterministically through a bundled Node script. A SessionStart hook injects the index as context at session start so Claude navigates straight to the right files instead of grepping. Incremental updates re-scan only what changed.
claude plugin install code-atlas@mistervitopro-plugin-marketplace
/code-atlas:map # full first-time scan
/code-atlas:update # incremental refresh
/code-atlas:query "what calls AuthService.login?"
Take a Markdown implementation plan, run it through a parallel agent swarm.
The analyzer breaks your plan into file-disjoint waves (≤6 agents per wave). Dev agents implement tasks in parallel; verifier agents check acceptance criteria per wave; the aggregator dedupes bugs and emits a fix-plan.md for re-runs. Per-wave git commits keep history bisectable. Optional TDD red-green mode.
claude plugin install plan-runner@mistervitopro-plugin-marketplace
/plan-runner:run path/to/implementation-plan.md
Consolidates scattered specs into a single canonical tree, and stubs out the docs you forgot to write.
adopt reorganizes spec files in place, grouped by module (multi-module repos) or feature (single-module). rewrite collapses everything to one file with optional source cleanup. The surface scanner walks your code and appends stubs for undocumented agents, skills, CLIs, and configs so nothing slips through.
claude plugin install jupiter@mistervitopro-plugin-marketplace
/jupiter:adopt # reorganize specs in place
/jupiter:rewrite # collapse to single master file
Vulnerability-aware dependency upgrade orchestrator across 7 ecosystems.
npx claudepluginhub mistervitopro/qa-claude-market --plugin plan-runnerConsolidate scattered specs into a canonical master-spec tree with a navigable index, and append stubs for undocumented public surface found in code
AI-powered code quality analyzer: specialized agents find security, performance, architecture, and correctness issues, then fix them via TDD
Generate square pixel-art sprites from semantic JSON grids + palettes. Deterministic, no external image model.
Scan a codebase and generate a comprehensive architecture index with graph-schema and query capability — directory map, key files, tech stack, patterns, dependencies, build commands, and queryable knowledge graph
Vulnerability-aware dependency upgrade orchestrator. Detects outdated packages across npm, Python, Go, Rust, Java, Kotlin, and C# ecosystems; recommends the safest-yet-most-recent version per package using OSV.dev; executes upgrades wave-by-wave with build/typecheck/test verification and clean git rollback on failure.
Harness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.