Menü
Migrate legacy AI-DLC intents to H·AI·K·U format
Mit Codex oder Claude installieren Kopieren Sie diesen Prompt, fügen Sie ihn in Codex, Claude oder einen anderen Assistant ein und lassen Sie die Skill-Seite prüfen und installieren.
Quick mode for small tasks — single-stage intent with auto-advance
Start a new H·AI·K·U intent — describe what you want to accomplish and the system creates a structured lifecycle for it
Full autonomous workflow — elaborate, plan, build, review, and deliver in one command
Open the SPA review pane for an intent so the user can browse units, feedback, stage artifacts, and outputs. Non-blocking on the workflow — leaves feedback that the engine picks up on the next tick. Use whenever the user wants to "see," "open," "show," or "look at" the current intent or a specific stage.
Reverse-engineer an existing feature into H·AI·K·U intent artifacts
Zero-ceremony single-unit execution — run one task directly through a stage's hat loop without intent or workflow scaffolding
Basierend auf der SOC-Berufsklassifikation
| name | haiku-migrate |
| description | Migrate legacy AI-DLC intents to H·AI·K·U format |
Convert legacy .ai-dlc/ intents to .haiku/ format.
Never apply a migration without showing a dry-run first. Bare haiku migrate is rejected by the binary, but the rule still holds when you pass --all or a specific slug: dry-run, get the user's OK, then apply.
ls .ai-dlc/. If the user named a slug, use that. Otherwise show the list and ask which one(s).haiku migrate <slug> (dry-run is the default). Show the user the output — what would be written, where, how many files.haiku migrate <slug> --apply. One slug at a time unless the user explicitly approved --all./haiku:haiku-pickup <slug> to continue execution.--apply — actually write. Default is dry-run.--all — migrate every intent in .ai-dlc/. Pair with --apply to commit.--force — re-migrate intents that already exist under .haiku/. Use sparingly.--allow-dirty — skip the git-clean precheck. Don't pass this without user approval; a dirty tree means the migration output gets tangled with unrelated in-progress work.Bare haiku migrate --apply rewrites every intent in .ai-dlc/ at once. In a monorepo, one such commit shows up in every open MR. The dry-run-then-confirm flow exists so this can't happen by accident.