Skills for using Animus — an autonomous agent orchestrator for software development workflows
Run and inspect Animus agent executions, direct provider runs, agent control, status, project-scoped agent memory, and agent message channels.
Product lifecycle agents — product owner, architect, auditor, docs-writer, devops, researcher personas
Guide a project from idea to autonomous engineering setup — interview the user, write VISION.md, AGENT_PRINCIPLES.md, registry, agents/workflows/phases/schedules YAML, scripts, and a first runnable task. Use when standing up Animus in a new project beyond the minimal /animus-setup scaffold.
Animus project config, daemon config, plugin config, agent runtime, environment variables, and state layout
Start, stop, restart, monitor the Animus daemon — health checks, events, logs, pool sizing, common issues
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.
The fastest way to turn Claude Code (or Codex, OpenCode, Cursor) into an autonomous engineering team that ships PRs while you sleep.
Animus is a Rust agent orchestrator. A daemon dispatches workflows from a queue, uses provider and subject plugins to run agents in isolated worktrees, runs your tests, opens PRs, and reviews them — on cron, on demand, or in response to events. animus-skills is the skill bundle that teaches your agent how to drive it.
You point a fresh agent at this README. It installs the CLI, links the skills, wires up MCP, and you start running workflows in about a minute.
Who this is for:
cd into any repo and run /animus-setup — scaffolds .animus/, writes a first workflow, starts the daemon/animus-getting-started — creates your first task and runs it end-to-endanimus daemon stream — watch the agent work in real timeRequirements: Claude Code (or Codex / OpenCode / Cursor), Git, bash, curl. Animus itself is a single Rust binary — no Node, no Python, no Docker.
Open a fresh Claude Code session and paste this. The agent does the rest — clones the repo, installs the animus CLI, links every skill into your host's skill directory, and writes .mcp.json if you're inside a project:
Install Animus + Animus Skills: run
git clone --single-branch --depth 1 https://github.com/launchapp-dev/animus-skills.git ~/.claude/skills/animus-skills && cd ~/.claude/skills/animus-skills && ./setup, then add an "Animus" section to CLAUDE.md (or AGENTS.md for Codex) listing the available slash commands:/animus-setup,/animus-bootstrap,/animus-getting-started,/animus-mcp-setup,/animus-workflow-authoring,/animus-pack-authoring,/animus-skill-authoring,/animus-troubleshooting. Restart the agent so the new MCP server (animus) is picked up.
For Codex CLI, swap the clone path to ~/.codex/skills/animus-skills and AGENTS.md instead of CLAUDE.md.
git clone https://github.com/launchapp-dev/animus-skills.git ~/animus-skills
cd ~/animus-skills && ./setup # auto-detects installed hosts
./setup --host claude # only Claude Code
./setup --host codex # only Codex
./setup --host all # every supported host (claude, codex, opencode, cursor, slate, kiro)
./setup --no-cli # skip the animus CLI install (use existing binary)
./setup --no-mcp # skip writing .mcp.json into the current project
The script:
animus CLI to ~/.local/bin/animus if it's not already on PATH.~/.claude/skills/<skill-name>, ~/.codex/skills/<skill-name>). Hosts discover skills flat as <host-skills-dir>/<name>/SKILL.md, so the script links one level down rather than the whole repo..mcp.json exposing the animus MCP server (skipped if one already exists or you pass --no-mcp)./plugin marketplace add launchapp-dev/animus-skills
This skips the CLI install and MCP wiring — pair it with a manual curl -fsSL https://raw.githubusercontent.com/launchapp-dev/animus-cli/main/scripts/install.sh | bash if you don't already have animus.
You: /animus-setup
Claude: [scaffolds .animus/, writes config.json, generates a first workflow YAML,
starts the daemon, verifies MCP is reachable]
You: Create a task to add rate limiting to the /api/upload endpoint.
Claude: [calls animus.subject.create with kind=task, then enqueues TASK-001]
You: Run it.
Claude: [calls animus.workflow.run — daemon picks up TASK-001, spawns a
Claude session in an isolated worktree, runs the standard pipeline:
research → plan → implement → test → review → PR]
You: animus daemon stream
[live JSONL — phase transitions, model calls, test runs, git ops]
You: Now schedule a nightly retro across all my repos.
Claude: [edits .animus/workflows/retro.yaml, adds a cron schedule, restarts daemon]
Every step is replayable from the queue — pause, resume, drop, reorder, hold. Failed phases are sticky (the daemon won't silently retry and burn API credits). Outputs land in ~/.animus/<repo-scope>/runs/ so you can audit exactly what each model did.
npx claudepluginhub launchapp-dev/animus-skills --plugin animus-skillsReliable 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
Core skills library for Claude Code: TDD, debugging, collaboration patterns, and proven techniques
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
Design fluency for frontend development. 1 skill with 23 commands (/impeccable polish, /impeccable audit, /impeccable critique, etc.) and curated anti-pattern detection.
Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls