plan-to-task-list-with-dag

Installation
SKILL.md

Non-negotiable rules:

  1. Start with scope challenge and mode selection using AskUserQuestion.
  2. Explore existing code before decomposing. Never invent file paths.
  3. Build one canonical plan object, then emit both markdown and JSON from it.
  4. Every task must have explicit ownership, validation, dependency handling, and an Agent field.
  5. Keep the skill body focused on workflow. Load plan schema, agent mapping, and examples from references.
  6. Every task needs 2-3 testable acceptance criteria, at least one covering a failure or edge case.
  7. Never create tasks touching more than 3 files, never create circular dependencies, never assume codebase structure.
  8. If a task claims side effects (persistence, WAL, network I/O, registration), state where the capability comes from.
  9. For new files, explicitly assign export/registration/wiring ownership to a task.
  10. If a task can be "completed" with placeholders or dead wiring, split semantic hardening into an explicit follow-up.
  11. Treat JSON as the source of truth. Render markdown from the same canonical plan object.
  12. Do not use vague contract language ("internal update", "eventually skipped", "graceful degradation") without defining owner, behavior, and recovery.
  13. Distinguish local reuse from external references. Do not present external docs or clones as local code.
Related skills
Installs
65
Repository
ulpi-io/skills
First Seen
Mar 15, 2026