A plugin marketplace for Claude Code. Browse, install, and manage plugins for development workflows and AI code planning and reviews.
Register the marketplace in Claude Code:
/plugin marketplace add jpeggdev/buildwithjpeggThen install any plugin:
/plugin install development-workflowA complete software development workflow built on composable skills and automatic invocation. Your coding agent steps back to understand what you're building before writing code, produces a reviewable design, then works through implementation with TDD, code review, and CI/CD built in.
Because skills trigger automatically, you don't need to do anything special -- describe what you want to build and the workflow activates.
The basic flow:
- evaluate -- Refines ideas through questions, explores alternatives, presents design for validation
- blueprint -- Breaks work into bite-sized tasks with exact file paths and verification steps
- delegate/build -- Dispatches subagents per task with two-stage review, or executes in batches with human checkpoints
- test-first -- Enforces RED-GREEN-REFACTOR for every implementation task
- seek-review -- Reviews against plan, reports issues by severity
- wrap-up -- Verifies tests, presents merge/PR/keep/discard options
Also includes: root-cause debugging, CI monitoring (ci-loop), stacked PR management (draft-prs, pr-stack), automated releases (auto-release), skill creation (craft-skill), and onboarding.
Components: 17 skills, 3 slash commands, 1 agent (code-reviewer), 8 rules, session-start hook
Also supports Codex and OpenCode -- see plugin README for setup.
Multi-agent orchestration framework for Claude Code, Codex CLI, and Gemini CLI. Provides 18 specialized agents, a 9-phase development workflow, and quality gates that enforce TDD, coverage thresholds, and spec-driven development.
The pipeline:
- Research -- Gather context and requirements
- Brainstorm -- Multi-agent debate to explore design alternatives
- Design Review Gate -- 5-agent parallel review (PM, Architect, Designer, Security, CTO)
- Plan -- Break work into implementation units
- Plan Review Gate -- 3 adversarial reviewers (Feasibility, Completeness, Scope)
- Execute -- 4-phase loop per work unit (Implement, Validate, Adversarial Review, Commit)
- Final Review -- Cross-check against requirements
- Self-Reflect -- Extract learnings into knowledge base
- Ship -- PR creation with coverage enforcement
Also includes: intelligent agent selection with decay scoring, layered config (tribunal.yaml), cross-CLI support, migration tools, and diagnostic checks.
Components: 16 skills, 19 commands, 18 agents, 8 rubrics, session-start hook
See tribunal repo for setup and usage.
MIT -- see LICENSE for details.