agentleFS
Sign inStart free
lulufoolulufoo/lulu-workflow/lulu-tasks/SKILL.md

lulu-tasks

Skill3 starsChanged 6 days agoMITView on GitHub

Split a delivered plan or approach decision into an independently executable task-dependency list.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    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:

    Do NOT proceed until you have read ../_runtime.md and loaded:

    line 20
  2. Very specific:

    Run $TT_START. It reads the delivered plan, or the delivered approach decision when no plan is delivered, from the cycle's delivered refs.

    line 40
  3. Very specific:

    Run $TT_CTX and pin the JSON as $CTX.

    line 41
  4. Very specific:

    When a unit returns, run $TT_CTX again and load the unit it names.

    line 51

≈ “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

tasks-workflow

Split a delivered plan, or an approach decision when no plan is delivered, into an independently executable task-dependency list. tech-doc in this skill means that reference document.

A task has kind: coding or kind: action.

  • coding changes code. Signatures, TDD order, and tdd_exempt apply only to this kind.
  • action reaches a stated goal by any means. It declares effects, lists acceptance criteria that evidence can answer one by one, and binds a worktree only when it names one. It does not commit code.

Delivery starts lulu-exec for the whole work order. That session dispatches by kind.

<HARD-GATE> Do NOT proceed until you have read `../_runtime.md` and loaded:
  • $SKILL_ROOT, $WORKFLOW_DIR, $PLATFORM, $CACHE_DIR from ## Platform Context
  • $SKILL_DIR = $SKILL_ROOT/lulu-tasks
  • Feature identification from ## Session Foundation</HARD-GATE>

Script Macros

MacroCommand
$TT_STARTpython3 "$SKILL_DIR/scripts/tt_start.py" --cycle-id "$CYCLE_ID"
$TT_FLOWpython3 "$SKILL_DIR/scripts/tt_workflow_control.py" --cycle-id "$CYCLE_ID"
$TT_CTX$TT_FLOW resolve-context

Subcommand contracts: module docstring / --help.

Start

  1. Identify the active cycle through _runtime.md § Session Foundation. Do not run $TT_START until $CYCLE_ID is confirmed.
  2. Run $TT_START. It reads the delivered plan, or the delivered approach decision when no plan is delivered, from the cycle's delivered refs. A non-zero result is Blocking.
  3. Run $TT_CTX and pin the JSON as $CTX.

Router

Load the unit named by $CTX.unit:

  • drafting → $SKILL_DIR/references/drafting.md
  • evaluating → $SKILL_DIR/references/evaluating.md
  • delivery → $SKILL_DIR/references/delivery.md

When a unit returns, run $TT_CTX again and load the unit it names. current_state Delivered: stop.

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.

Check your setup, free