Infinite Angel — daemon-supervised coding angels that never stop.
npx claudepluginhub jinto/infinite-angelInfinite Angel — coding angels that never stop. 12 skills for the full dev lifecycle.
Daemon-supervised coding agents that never stop.
Launch an agent. Go to sleep. Wake up to working code.
Quick Start • Skills • Pipeline • Acknowledgements
# 1. Install binaries
curl -sSL https://raw.githubusercontent.com/jinto/infinite-agent/main/install.sh | sh
source ~/.zshrc # or open a new terminal
# 2. Setup (Claude Code hooks + MCP)
ina setup
# 3. Install skills (in Claude Code)
/plugin marketplace add https://github.com/jinto/infinite-agent
/plugin install ina
# 4. Start daemon (pick one)
ina install # recommended: auto-start on login (macOS launchd)
ina daemon # or: run in foreground
# 5. Use
/ina:autopilot Add user authentication
Other Claude Code plugins are prompt libraries. ina is infrastructure.
| Prompt libraries | ina | |
|---|---|---|
| Agent crashes | Gone forever | Daemon restarts + resumes |
| Context resets | Start over | Pipeline state preserved |
| Long-running tasks | Hope for the best | Monitored + alerts |
| Multiple agents | Manual | Registry + coordination |
THINK PLAN BUILD REVIEW SHIP
┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐
│ Idea │ ──▶ │ Plan │ ──▶ │ Code │ ──▶ │ QA │ ──▶ │ PR │
│ Spec │ │ Tasks│ │ Test │ │ Fix │ │ Merge│
└──────┘ └──────┘ └──────┘ └──────┘ └──────┘
| Skill | Description |
|---|---|
autopilot | Full pipeline: think → plan → build → review → commit |
think | Idea → spec (technical / business / improve) |
plan | Consensus planning + TDD task breakdown |
build | Execute tasks (direct / subagent / team) |
review | Code review + fix-first auto-correction |
research | Multi-angle decomposition + parallel search |
design | UI implementation + visual verification |
test | Test runner + failure analysis + fix |
ship | Create PR with auto-generated summary |
guard | Safety guardrails for unattended execution |
Don't know which skill to use? Just describe what you want — ina auto-selects the right skill via the built-in reference guide.
/ina:think I want to add user authentication
Socratic interview → multi-perspective validation (Architect/Critic/CEO) → spec document.
/ina:plan .omc/specs/think-auth.md
Consensus planning (Planner → Architect → Critic) → TDD task breakdown → TASKS.md.
/ina:build
Auto-delegates: direct execution for 1 task, subagent parallelism for 2-3, team for 4+.
/ina:autopilot Add user authentication with OAuth2
Full pipeline: think → plan → build → review → commit. Crash-recoverable via .state/pipeline.json.
/ina:review
External code review (Codex) + fix-first auto-correction + loopback protocol.
/ina:test
Root cause analysis + fix + re-run (max 5 cycles).
/ina:ship
Auto-generates summary from git log/diff + runs tests before PR creation.
autopilot: think → plan → build → review → commit
↑ │
└─────────┘ (loopback, max 3)
Crash recovery via .state/pipeline.json — the daemon restarts the agent and resumes from the recorded stage.
Multi-perspective validation — think and plan invoke Architect, Critic, and CEO subagents in parallel to validate specs and plans before execution.
Execution delegation — build auto-selects direct execution, subagent parallelism, or team coordination based on task independence.
┌─────────────────────────────────────────┐
│ Skill Layer (SKILL.md) │
│ In-session orchestration via │
│ Claude Code native tools │
│ │
│ autopilot / think / plan / build / │
│ review / test / ship / guard │
├─────────────────────────────────────────┤
│ Daemon Layer (Go binary) │
│ Out-of-session process supervision │
│ Crash recovery, restart, Discord alerts│
│ │
│ ina daemon / watcher / hooks / MCP │
└─────────────────────────────────────────┘
go build -o ina .
go build -o ina-mcp ./mcp/
go test ./... -count=1 -race
# E2E skill routing tests (costs API credits)
INFA_E2E=1 go test ./test/ -run TestSkillRouting -v
Official prompts.chat marketplace - AI prompts, skills, and tools for Claude Code
Open Design — local-first design app exposed to coding agents over MCP. Install once with your agent's plugin command and projects/files/skills are reachable through stdio.
Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations