continuous-agent-loop

Runs an AI agent repeatedly with safety checks and recovery controls.

Installation

Run `npx skills add "https://github.com/affaan-m/ecc" --skill "continuous-agent-loop"` to install this skill, then follow its SKILL.md instructions for my next request.

Paste this into Claude Code, Cursor, or any agent that can run commands.

What this skill does

What it does: This skill runs a repeating loop where an AI agent works on a task, checks if the work is good enough, and tries again if needed. It includes safety checks to stop the loop if something goes wrong, like if the agent keeps making the same mistake or costs get too high. When to use it: Use this when you need an AI to keep working on a problem until it is solved, like fixing code or writing documents. It is helpful when the task might need several tries to get right. You would use it for important work where you want the AI to keep improving until the result is good.

SKILL.mdShow the author's original SKILL.md

---
name: continuous-agent-loop
description: Patterns for continuous autonomous agent loops with quality gates, evals, and recovery controls. Use when running an agent loop that must self-check, gate on evals, and recover from failures.
metadata:
  origin: ECC
---

# Continuous Agent Loop

This is the v1.8+ canonical loop skill name. It supersedes `autonomous-loops` while keeping compatibility for one release.

## Loop Selection Flow

```text
Start
  |
  +-- Need strict CI/PR control? -- yes --> continuous-pr
  |
  +-- Need RFC decomposition? -- yes --> rfc-dag
  |
  +-- Need exploratory parallel generation? -- yes --> infinite
  |
  +-- default --> sequential
```

## Combined Pattern

Recommended production stack:
1. RFC decomposition (`ralphinho-rfc-pipeline`)
2. quality gates (`plankton-code-quality` + `/quality-gate`)
3. eval loop (`eval-harness`)
4. session persistence (`nanoclaw-repl`)

## Failure Modes

- loop churn without measurable progress
- repeated retries with same root cause
- merge queue stalls
- cost drift from unbounded escalation

## Recovery

- freeze loop
- run `/harness-audit`
- reduce scope to failing unit
- replay with explicit acceptance criteria

Mirrored from the author's public source. Install counts from the open skills registry.