research-orchestrator-skill
Orchestrate hypothesis-driven research across one or many agent sessions using a scored PLAN queue, shared DISCOVERIES that a different host cross-checks once, adaptive workers, CPU/GPU/Other resource routing, and a durable HANDOFF log. Use when starting or resuming research, coordinating parallel experiments, prioritizing hypotheses, cross-checking conclusions from another host such as Codex or Claude Code, scaling work to available compute, or preserving project continuity across sessions.
Pinned to revision 572b37475b9f, so it is the text this page describes rather than whatever the author pushed since.
Files
- skills/research-orchestrator-skill/SKILL.md
- skills/research-orchestrator-skill/agents/openai.yaml
- skills/research-orchestrator-skill/assets/icon.svg
- skills/research-orchestrator-skill/scripts/check_project.py
- skills/research-orchestrator-skill/scripts/init_research_orchestrator.py
- skills/research-orchestrator-skill/templates/AGENTS.md.template
- skills/research-orchestrator-skill/templates/DISCOVERIES.md.template
- skills/research-orchestrator-skill/templates/HANDOFF.md.template
- skills/research-orchestrator-skill/templates/PLAN.md.template
Every link opens the file at its source, pinned to the revision this page describes.