agentleFS
markoblogomarkoblogo/abvx-agent-skills/skills/happy-handoff/SKILL.md

happy-handoff

Skill16 starsChanged 5 months agoMITView on GitHub

Continue the current Codex Desktop task in Happy on Android when the user says "ухожу", asks to move or hand off to Happy, or invokes this skill explicitly.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Never add --yolo, bypass approval settings, or expand permissions.

    line 33
  2. Distinctive:

    Do not claim that Android received the task until the user observes it there.

    line 34
  3. Distinctive:

    Do not enable voice or daemon autostart as part of a handoff.

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Happy Handoff

Resume the same Codex task in Happy after reaching a safe checkpoint. The handoff request authorizes starting Happy for the current task. It does not authorize commits, pushes, deployments, destructive actions, or unrelated changes.

Workflow

  1. Finish the current tool call or other indivisible operation.
  2. Preserve the workspace. Do not clean, stash, commit, or rewrite dirty work for the handoff.
  3. Capture a compact checkpoint: completed work, current branch or directory, validation state, dirty files, active processes, and the next concrete step.
  4. Require CODEX_THREAD_ID, the current working directory, an authenticated Happy CLI, and the system script command. If a prerequisite is unavailable, report the exact blocker instead of opening an unrelated session.
  5. As the final tool action, run:

    python3 "${CODEX_HOME:-$HOME/.codex}/skills/happy-handoff/scripts/launch.py"
  6. Immediately return the checkpoint so it is stored in the task before Happy resumes it. Say that the handoff is scheduled and ask the user to open Happy on Android.

The launcher waits briefly, then starts happy codex --resume for the exact task in the same directory. Use --dry-run to validate prerequisites without scheduling a session.

Guardrails

  • Never add --yolo, bypass approval settings, or expand permissions.
  • Do not claim that Android received the task until the user observes it there.
  • Do not enable voice or daemon autostart as part of a handoff.
  • Starting the Happy daemon when it is needed is allowed by the handoff request.
  • Windows is currently unsupported; macOS and Linux use different script syntax and are handled explicitly.

Output Shape

Report:

  • checkpoint;
  • task ID and workspace;
  • launcher status: dry-run, scheduled, or blocked;
  • next step in Happy;
  • unverified boundary: Android receipt until observed.

Provenance

Original ABVX workflow around the MIT-licensed slopus/happy CLI. Happy is an external companion and is not bundled with this repository.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.