Practical patterns, starters & CLI tools for loop engineering with AI coding agents. Design systems that prompt and orchestrate agents (inspired by Addy Osmani and Boris Cherny). Includes loop-audit, loop-init, loop-cost.
TypeScript agentic-aiai-agentsai-codinganthropicautomationclaudeclaude-codecodexcoding-agentsdevops-automationdevtoolsgithub-actionsgrokllmloop-engineeringmcpprompt-engineering
Goal
Harden tools/loop-action/action.yml so inputs.command is not expanded unquoted into the shell (fragile for multi-arg / multi-line agent invocations).
Files
tools/loop-action/action.ymltools/loop-action/README.md(document the safe usage)
Acceptance criteria
- Sandbox and direct paths pass the command safely (e.g. write to a temp script, or use
bash -lcwith a properly quoted string / env var) - Multi-line
command: |examples inexamples/github-actions/*.ymlstill work - README warns against embedding untrusted input in
command - No behavior change for the simple placeholder
echoexamples beyond safer invocation
Estimated time: ~45–60 minutes (tooling; still scoped)
Comment "I'll take this" to get assigned.