Autonomous Loops
Loop patterns for autonomous Claude Code: sequential pipelines, infinite agentic loops, PR loops, and RFC-driven multi-agent DAGs. Compatibility version.
- What
- Loop patterns for autonomous Claude Code: sequential pipelines, infinite agentic loops, PR loops, and RFC-driven multi-agent DAGs. Compatibility version.
- Cost
- Free
- Needs
- Use "Autonomous Loops" with your Muse.
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor: patterns, architectures and reference implementations for running Claude Code autonomously in loops, from the simplest to the most sophisticated. The spectrum covers sequential pipelines of non-interactive claude -p calls for scripted daily work; the NanoClaw REPL for interactive persistent sessions; infinite agentic loops for parallel content generation and spec-driven work; continuous PR loops for multi-day iterative projects with CI gates; the de-sloppify cleanup pass after any implementer step; and Ralphinho-style RFC-driven multi-agent DAG orchestration with a merge queue for large features. Covers choosing the right loop architecture for the problem, context persistence across iterations, parallel agents with merge coordination, and quality gates with cleanup passes. From the affaan-m/ECC repository (MIT). Honest caveats: retained for compatibility only, the canonical skill is now continuous-agent-loop, so prefer that for new work; autonomous loops multiply token spend and need stop conditions plus human oversight. Skill Harbor never reviews the code, review it yourself before use.
Version:
Install
Copy the install package below, then paste it into MuseThe install prompt below already includes the vetting steps: your agent follows the community checklist before installing anything with executable code. Want more?
Use "Autonomous Loops" with your Muse. Compatibility note: this skill is retained for one release; the canonical skill is now continuous-agent-loop. Use this copy only to keep existing workflows working. 1. Open the skill: https://github.com/affaan-m/ECC/blob/main/skills/autonomous-loops/SKILL.md and copy the full SKILL.md text. 2. Paste it into a chat with Muse and add: "Design a sequential claude -p pipeline for [daily task]" or "Set up a Ralphinho-style RFC-driven loop for [feature]." 3. Define the stop condition and the quality gate before the first iteration; a loop without an exit is a billing event. Tip: start with the simplest pattern that could work; move up the spectrum only when the simple loop provably cannot drive the task. Safety: a skill is plain-text instructions; it runs nothing by itself. Autonomous loops spend tokens while you sleep; cap iterations and keep destructive steps behind human approval.
Saved to your recent installs. Find it anytime on /connect.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.