Autonomous Loops Patterns
Design and run unattended Claude Code loops, from simple claude -p pipelines to multi-agent DAG systems
- Repository
- affaan-m/ECC
- Stars
- 295,127
- Licence
- MIT
- Files
- 1 file
What it does
This skill gives the agent a reference guide for setting up Claude Code to run repeatedly and automatically without a human in the loop. It explains a spectrum of loop designs, from a simple ordered sequence of non-interactive claude -p calls to large RFC-driven multi-agent orchestration, and includes an example bash script that chains implement, cleanup, verify and commit steps. It helps the agent choose an architecture that fits a given problem and apply design principles such as isolated steps, order dependence and exit-code propagation. The file is also a compatibility note: the skill is kept for one release, and the canonical name is continuous-agent-loop.
Great for
- Developers who want to automate repetitive coding steps in CI or scripts
- Teams deciding whether a simple pipeline or a multi-agent setup suits a large feature
- Engineers building continuous development pipelines with quality gates
Skip it if
- Non-developers, since the content assumes familiarity with shell scripts and Claude Code CLI; running the example unattended can change many files and commit code without review
- Users looking for the current loop guidance, because the file says new guidance belongs in continuous-agent-loop
- Tasks needing the model routing or advanced features, because the file defers those to external documentation not included here
Try asking
- “Help me design an unattended Claude Code pipeline that implements a feature, cleans up, tests and commits”
- “Which loop pattern should I use for a large feature with several parallel agents and a merge queue?”
- “Write a daily-dev.sh script that chains claude -p steps with set -e so failures stop the pipeline”
- “Explain the De-Sloppify pattern and where to add a cleanup pass after implementation”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
autonomous-loops.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip autonomous-loops.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip autonomous-loops.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Review the generated loop script and run it first in a sandbox or a throwaway branch before allowing unattended commits
- Prefer positive instructions in each step; the file advises against negative instructions and recommends a separate cleanup step instead
- Use the pattern table to start with the simplest loop type that solves the problem
What's inside
- SKILL.md
Redistributed unmodified from affaan-m/ECC under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.