DEDUPED Hot AI repos, turned into use cases Get 5 a day

How to turn agent requests into numbered steps you can follow with Open Steps

When the agent needs you to run, paste, click or approve something, get exact numbered steps instead.

Who it is forA non-technical operator who has to run commands or click through screens for an AI agent that is working on a project.

The problemWhen an agent needs a human action, the instructions are often vague or assume engineering knowledge, so the operator stalls or makes mistakes.

The outcomeNumbered steps that a non-technical person can follow, with the agent trying everything it can itself first and asking only for what truly needs the person.

What you need first

  • git, to download and update the pack
  • One of Claude Code, Codex CLI, Cursor CLI or Gemini CLI installed
  • gh, GitHub's command-line tool (optional; without it some output says "not checked")

Steps

  1. Run git clone https://github.com/kharmanskyi/open-steps.git
  2. Run mkdir -p ~/.agents/skills && cp -R open-steps/skills/os-* ~/.agents/skills/ on Codex CLI, Cursor CLI or Gemini CLI
  3. Add the routing block from open-steps/docs/routing-block.md to your tool's standing instructions file
  4. Run bash open-steps/doctor.sh to check that the install is wired (inferred, check the README)
  5. When the agent asks you to act, check that the answer uses numbered steps

Prompt for your agent

In the open-steps repo at <repo-path>, use the os-step-by-step skill for the next action that needs a human on <project-name>. Expected output: numbered steps a non-technical person can follow, with only the actions that truly need the person listed. Verify that the agent tried every step it could do itself before asking.

Caveat. The README notes that the Codex CLI, Cursor CLI and Gemini CLI setup has not been tested for Claude-only permission features, so expect the ask-first behavior on those tools to differ.

Share: X LinkedIn Hacker News Email Copy link

Disclaimer

Deduped lists open-source repositories that belong to other people. We do not own, maintain, audit, or vouch for any of them, and a repo appearing here is not an endorsement or a safety check. The use cases, steps, and prompts are written by AI and may be incomplete, outdated, or just wrong. Where a use case is marked "Seen in the wild", we link to public posts where someone described something similar, but we have not verified those posts and the use case itself is still our AI's own write-up. Use all of this at your own risk. Only run code, commands, or agents on systems and accounts you are authorized to use, and read the code and every command before you run it, especially anything that touches credentials, files, or the network. Nothing here is professional, security, legal, or financial advice. All trademarks and project names belong to their respective owners, and their use here does not imply affiliation. Content is provided as is, without warranties of any kind. Found a mistake, a safety problem, or something you want removed? Email [email protected] and we will look at it promptly. Full disclaimer page