agentleFS
ItsWendellItsWendell/palot/.agents/skills/palot-desktop-dev/SKILL.md

palot-desktop-dev

Skill194 starsChanged 46 days agoMITView on GitHub

Operate Palot's Electron build and channel lifecycle. Use for main, preload, or lifecycle changes; dev supervisor or launcher work; native dependencies or generated binaries; Dev/Nightly channel configuration; packaging or installation; and start, restart, build, or channel failures. Use test-palot-desktop for native UI QA.

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 3 are distinctive, and 11 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 11

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Keep the OpenCode task host stable while the development client changes.

    line 8
  2. Very specific:

    Use installed Palot Nightly as the host when changing Palot itself.

    line 12
  3. Very specific:

    Use bun run dev for live integration.

    line 14
  4. Very specific:

    Use bun run dev:visible when rendering, interaction, accessibility, screenshots, or visual inspection require a visible native window.

    line 15
  5. Very specific:

    Use bun run dev:focus only when focus behavior matters or the user explicitly asks for a foreground launch.

    line 16
  6. Very specific:

    Read repository-root ../../../docs/desktop-testing.md before adding an E2E scenario or using real credentials.

    line 27
  7. Very specific:

    Start the supervisor with bun run dev from the repository root for background integration.

    line 31
  8. Very specific:

    Run bun run dev:info in that worktree to get its renderer URL, Electron PID, CDP port, and data root.

    line 32
  9. Very specific:

    After native dependency or generated-binary changes, run bun run dev:restart -- "reason" as the final tool call.

    line 34
  10. Distinctive:

    After supervisor or control-script changes, stop its terminal with Ctrl-C and start it again.

    line 35
  11. Very specific:

    Run bun run package:nightly:mac to create channel-specific artifacts.

    line 41
  12. Very specific:

    Run bun run install:nightly:mac to build, ad-hoc sign, verify, install, and launch /Applications/Palot Nightly.app.

    line 42
  13. Distinctive:

    Leave any previous Nightly installation moved to Trash available for recovery.

    line 45
  14. Very specific:

    Keep validation proportional to the changed behavior.

    line 49

≈ “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

Palot Desktop Development

Keep the OpenCode task host stable while the development client changes.

Choose The Loop

  • Use installed Palot Nightly as the host when changing Palot itself.
  • In agent worktrees, default to the smallest matching bun run test:e2e -- <scenario>. Add --visible only when the user wants to watch it.
  • Use bun run dev for live integration. It connects to the shared user OpenCode service and starts hidden so it does not cover or focus another app.
  • Use bun run dev:visible when rendering, interaction, accessibility, screenshots, or visual inspection require a visible native window. It does not take keyboard focus.
  • Use bun run dev:focus only when focus behavior matters or the user explicitly asks for a foreground launch.
  • Build Nightly and Dev against the exact declared @opencode/client version. Reuse healthy compatible stable 2.x services and reviewed beta services; an unreviewed beta needs exact-version consent. A different patch version is not permission to replace a service. Service replacement can interrupt active turns.
  • Existing user-installed OpenCode is preferred for confirmed local startup; packaged/app-cached binaries are fallbacks or an explicit user preference. Delegate confirmed package-manager updates to the official CLI, and keep executable updates separate from service restarts. Never update the task host as a side effect of implementing or testing these controls.

Read repository-root ../../../docs/desktop-testing.md before adding an E2E scenario or using real credentials.

Iterate

  1. Start the supervisor with bun run dev from the repository root for background integration. Use bun run dev:visible when the QA evidence requires a visible window, and bun run dev:focus only for focus-specific behavior or an explicitly requested foreground launch.
  2. Run bun run dev:info in that worktree to get its renderer URL, Electron PID, CDP port, and data root.
  3. Let Vite HMR apply renderer-only changes. Let the supervisor restart Electron after main or preload builds.
  4. After native dependency or generated-binary changes, run bun run dev:restart -- "reason" as the final tool call.
  5. After supervisor or control-script changes, stop its terminal with Ctrl-C and start it again. A running supervisor cannot replace itself.

Each live worktree gets dynamic renderer and CDP ports plus separate Palot state and logs. OpenCode sessions, config, auth, and models remain shared. Use only the ports returned by dev:info and target the owned terminal session for shutdown.

Install Nightly

  • Run bun run package:nightly:mac to create channel-specific artifacts.
  • Run bun run install:nightly:mac to build, ad-hoc sign, verify, install, and launch /Applications/Palot Nightly.app.
  • A successful routine install is complete when that command exits successfully. Trust its release and signature verification; report the installed path without repeating bundle, process, log, or UI checks.
  • Inspect identity, icon, signature, replacement behavior, launch, or OpenCode connection only when the related packaging code changed, the command failed, or the user requested package QA.
  • Leave any previous Nightly installation moved to Trash available for recovery.

Verify

  • Keep validation proportional to the changed behavior. Use focused checks before broad bun run check or full builds.
  • For desktop integration behavior, run the smallest matching E2E scenario.
  • For supervisor changes, confirm restart success, a new Electron PID, a live Vite process, native Liquid Glass, and restored task state.
  • Load test-palot-desktop only for native UI inspection, screenshots, or explicit smoke QA. Keep live QA read-only unless the user approved a disposable test task.

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.