Claude Skills by jleechanorg

Run the Dark Factory DOT pipeline runner against a goal.

Dark Factory spec workflow (`/factory-spec`, `/fs`): create a main spec and an attractor spec through the spec-generation pipeline, review an existing spec against acceptance criteria, or display the canonical reference for pipeline node graphs — gates, node types, edge conditions, handler mappings. Create mode runs `dark-factory --pipeline slim/spec_gen.dot` and must produce a codex-cold-reviewed `spec.md` and `attractor_spec.md`; review and show modes stay in-session.

Live Auto-Factory status audit — inspects real coding CLI conversations on jeff-ubuntu (/linux), captures tmux panes, audits evaluator/coder transcripts, verifies authentic test and commit progress, and reports 7-gate assessments.

End-to-end auto-factory driver — picks up beads + GH issues tagged factory, dispatches coders to drive worldai PRs to /green + /advice + /er, runs verifier ticks, repeats until done. Designed to work even when GH API is rate-limited (falls back to beads-only).

Common commands and CLI invocation reference for the Dark Factory DOT pipeline runner — running smoke and full gated pipelines, CXDB failure diagnosis with df-healer, pipeline graph visualization, and test suite execution.

Guide for implementing, registering, and testing new pipeline node types and handlers in the Dark Factory DOT runner.

Build a developer reproduction bundle for agent-review failures, missed-plan retrospectives, Claude/Codex transcript comparisons, or cases where Anthropic/OpenAI engineers need raw conversation JSONL, subagent state, repo evidence, hashes, and a concise replay guide.

Run matched A/B reviewer calibration for /f through the binary-owned review controller, using digest-bound prompts, envelopes, and receipts.