agentleFS
wado-langwado-lang/wado/.claude/skills/on-task-done/SKILL.md

on-task-done

Skill120 starsChanged 16 days agoMITView on GitHub

Task-completion flow: `/cr` (a `/code-review` answered with `/code-review-response`, which ends with `/distill`), update docs (spec/cheatsheet/compiler/optimizer), then run `mise run on-task-done` (build, clippy-fix, golden + format fixtures, doc-stdlib, format, tests; the longest task there is) and commit its generated changes. Invoke ONLY when the user asks for it by name or explicitly asks to run the completion flow — never on your own initiative, and not because a task looks finished.

The file

We split this skill into the rules it gives an agent

5 rules. About 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

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

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

    Don't pipe it into tail — you lose the full log.

    line 39
  2. Distinctive:

    Run it in the background and read the log when it lands.

    line 41
  3. Very specific:

    If the changes only include Wado programs, only mise run test-wado is required.

    line 42
  4. Very specific:

    If the changes only include documentation, only mise run format and mise run check are required.

    line 43
  5. Very specific:

    If cargo test segfaults, reduce parallelism:

    line 52

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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

Overview

First, review the branch with /cr: a review answered with /code-review-response, which ends with /distill.

Then update docs if applicable:

  • docs/spec-*.md
  • docs/cheatsheet.md
  • docs/compiler.md and docs/optimizer.md, only when what they map changes: a phase, pass, or IR, or a rule that holds across phases. How one works belongs in its module doc.

Then, run time mise run on-task-done, which includes:

cargo build
mise run clippy-fix
mise run update-golden-fixtures
mise run update-golden-format-fixtures
mise run doc-stdlib
mise run format
mise run test
mise run test-wado
mise run check
mise run check-grammar
mise run check-highlight

on-task-done is the longest task there is.

Notes

  • Don't pipe it into tail — you lose the full log. Redirect to a file and inspect it afterwards if you need to trim output.
  • The changes generated by the command are necessary. Commit them.
  • Run it in the background and read the log when it lands.
  • If the changes only include Wado programs, only mise run test-wado is required.
  • If the changes only include documentation, only mise run format and mise run check are required.

Flaky Commands

Some commands are flaky and may segfault:

  • mise run update-golden-fixtures — the golden-dump batch tool sometimes segfaults mid-run (pre-existing issue). If it fails, run the individual golden generation manually for any new fixtures, then proceed.
  • mise run test-e2e (i.e., cargo test --workspace --test e2e) — the e2e test runner sometimes crashes with SIGSEGV when running all tests in parallel (pre-existing issue unrelated to compiler correctness).

If cargo test segfaults, reduce parallelism:

cargo test --workspace --test e2e -- --test-threads=4

Or run individual fixture tests directly:

cargo test --workspace --test e2e -- <fixture_name>

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.