claude-corps

Parallel agentic development framework for Claude Code
Turn Claude Code into an autonomous development team — multiple agents working simultaneously on isolated branches, coordinated by an orchestrator, with structured workflow that improves across sessions.

Quick Start · Why? · Skills · Auto-Run · FAQ

---

Why claude-corps?

Claude Code is powerful on its own. claude-corps makes it a team.

• Parallel execution — Dispatch 3-5 Claude agents working simultaneously on different tasks, each in an isolated git worktree
• Sequential execution — Dependent tasks run one at a time on the branch via --sequential — no worktree overhead, no merge conflicts
• Full lifecycle coverage — From spec to dispatch to PR to code review, every step has a skill
• Autonomous multi-hour runs — /auto-run chains dispatch, reconcile, and repeat until your entire backlog is done
• Human in the loop — Agents execute, you decide. PRs are created, never auto-merged

---

Quick Start

Option A: Plugin Install

In Claude Code:

/plugin marketplace add josephneumann/claude-corps
/plugin install claude-corps@claude-corps

Note: Plugin install namespaces skills as /claude-corps:orient, /claude-corps:dispatch, etc. For the full un-namespaced experience with CLAUDE.md integration, use Option B.

Option B: Full Install (Recommended)

git clone https://github.com/josephneumann/claude-corps.git ~/Code/claude-corps
cd ~/Code/claude-corps && ./install.sh

This symlinks skills, agents, hooks, scripts, and docs into ~/.claude/ for un-namespaced /orient, /dispatch, etc.

Then in any project:

claude
> /orient              # Survey your project and identify parallel work
> /dispatch --count 3  # Spawn 3 workers in isolated worktrees
> /auto-run            # Or go fully autonomous

---

The Workflow

graph LR
    DP["/deep-plan"] --> PR["/product-review"]
    PR --> A["/spec"]
    A --> REV["Deep Reviews"]
    REV --> MR["/multi-review --plan"]
    MR --> DEC["Decompose"]
    DEC --> D["/orient"]
    A -.->|"or skip reviews"| D
    D --> E["/dispatch"]
    E --> F["Workers: /start-task"]
    F --> G["Workers: /finish-task"]
    G --> H["/reconcile-summary"]
    E -.->|"/auto-run loop"| H

Phase	What happens
Plan	/deep-plan orchestrates the full pipeline: /product-review challenges scope, /spec writes the plan, deep eng/design reviews add rigor, /multi-review --plan catches issues pre-coding, then decomposes to Linear. Or run each skill individually.
Execute	/orient surveys the project. /dispatch spawns workers — parallel (worktree-isolated, default) or sequential (--sequential, direct on branch). Each worker implements, tests, and writes a session summary. /auto-run does this in a loop until all tasks are done.
Review	/multi-review runs parallel specialized code review. /reconcile-summary syncs worker output with the task board.

---

Claude Code Skills Reference

All workflow capabilities are implemented as slash commands in skills/.

Planning

Skill	Purpose
/deep-plan	Full planning pipeline with checkpoints: product review, optional design exploration, spec, deep eng/design reviews, plan-stage multi-review, decomposition. Single entry point.
/design-shotgun	Generate 3-5 intentionally different UI directions before committing to a design approach
/product-review	Product-taste review with interrogation mode, assumption mapping, and devil's advocate challenges. EXPAND / HOLD / REDUCE / DESIGN modes
/spec	Research, plan, optionally decompose into Linear issues
/spec --deepen	Enhance an existing plan with parallel research

Execution