Autonomous Loops — Patterns for Claude Code Automation

Running Claude in loops enables spec-driven development, CI/CD-style pipelines, and iterative refinement without human intervention between steps.

Quick Start

Choose your pattern by complexity:

1. Sequential Pipeline (simple) — Chain claude -p calls for linear workflows
2. Persistent REPL (simple) — Interactive sessions with history
3. Spec-Driven Parallel (medium) — Deploy N agents from spec, manage waves
4. PR Automation Loop (medium) — PR creation, CI fix, auto-merge
5. De-Sloppify Pass (add-on) — Cleanup step after any implementation
6. RFC-Driven DAG (advanced) — Multi-unit parallel work with dependency graph

Pattern Spectrum

References

• references/sequential-pipeline.md — Basic claude -p loops with examples
• references/persistent-repl.md — NanoClaw-style session persistence
• references/parallel-agents.md — Spec-driven deployment with wave management
• references/pr-automation.md — Continuous Claude PR loop with CI gates
• references/de-sloppify.md — Quality cleanup pattern
• references/dag-orchestration.md — RFC-driven multi-unit coordination

Key Principles

1. Isolation — Each loop iteration gets fresh context (no bleed-through)
2. Context Persistence — Use files (SHARED_TASK_NOTES.md) to bridge iterations
3. Exit Conditions — Always set max-runs, max-cost, max-duration, or completion signal
4. No Blind Retries — Capture error context for next iteration
5. Separate Concerns — Different loop patterns for different problem sizes

Decision Matrix

Is this a single focused change?
├─ Yes → Sequential Pipeline
└─ No → Do you have a spec/RFC?
         ├─ Yes → Do you need parallel work?
         │        ├─ Yes → DAG Orchestration
         │        └─ No → PR Automation Loop
         └─ No → Do you need many variations?
                  ├─ Yes → Parallel Agents + Spec
                  └─ No → Sequential Pipeline + De-Sloppify

Anti-Patterns

❌ Infinite loops without exit conditions

❌ No context bridge between iterations

❌ Retrying the same failure without capturing error context

❌ Negative instructions instead of cleanup passes

❌ All agents in one context window (reviewer should never be the author)

❌ Ignoring file overlap in parallel work

Adapted from everything-claude-code by @affaan-m (MIT)