Light Orchestrator
Coordinate and recover multi-stage research projects with checkpoints and state management.
Installation
- Make sure Claude is on your device and in your terminal.
Skills load from
~/.claude/skills/when Claude Code starts up — so you need it on your machine first. If you don't have it yet, install it once with the command below, then runclaudein any terminal to verify.One-time setupnpm i -g @anthropic-ai/claude-codeAlready have it? Skip ahead.
- Paste into Claude Code or into your terminal.
This copies the whole skill folder into
~/.claude/skills/light-orchestrator-light0305/— the SKILL.md plus any scripts, reference docs, or templates the skill ships with. Safe default: works for every skill.Faster alternative (instruction-only skills)
Skips the clone and grabs only the SKILL.md file. Don't use this if the skill ships Python scripts, reference markdowns, or asset templates — they won't be downloaded and the skill will fail when it tries to load them.
Quick install (SKILL.md only)Sign up to copy - Restart Claude Code.
Quit and reopen Claude Code (or any other agent that loads from
~/.claude/skills/). New skills are picked up on startup. - Just ask Claude.
Skills auto-activate when your request matches the skill's description — no slash command needed. Trigger phrases live in the skill's own frontmatter; you can read them in the “What this skill does” section above.
Prefer to read the source first? Open on GitHub.
When Claude uses it
Coordinate and recover multi-stage Light research projects through the canonical .light/passport.yaml state, stages 1-13, resident overlays, checkpoints, findings, parallel joins, stale propagation, handoffs and user-authorized reroutes. Use for a new/resumed/partial/dirty/failed/stale/delivered research project; when the user says continue, resume, take over, checkpoint, reroute, recover or deliver; or when work crosses two or more Light research stages. Never turn frontend-design/system-design/patent-disclosure/software-copyright or overlays into stages, never execute a suggested back-edge without explicit user authorization, and never declare delivery from file existence alone.
What this skill does
Light Orchestrator manages multi-stage research projects through a canonical state file (.light/passport.yaml), enabling researchers to coordinate workflows, checkpoint progress, recover from failures, and deliver completed work. It enforces strict boundaries: never imposing research direction, silently rewriting user work, or claiming delivery without verification. The skill reads project intake, routes tasks through stages 1–13, tracks evidence states, manages parallel branches with joins, and handles reroutes only after explicit user authorization.
What it does
- Intake and state management — classifies projects as new, resumed, partial, dirty, failed, stale, or delivered; preserves the canonical passport and enforces schema migrations when needed.
- Stage routing — maintains a forward DAG from stage 1 (literature-search) through stage 13 (review-rebuttal), respecting dependencies, parallel branches, and joins; never conflates overlays or engineering skills with scientific stages.
- Checkpoint validation — runs
run_checkpoint.pyto record stage completion using producer findings, timestamps, and content hashes; reports critical, warn-only, or pass results without overwriting user work. - Reroute and recovery — interprets
reroute.pysuggestions (rework, admission-hold, known-limitation, manual) and executes back-edges only viapassport.py add-back-edgeafter explicit user authorization with evidence locators. - Stale propagation and handoff — tracks upstream byte changes that invalidate dependent stages, runs
stale-checkandverify-handoffto confirm integrity, and resumes from the reported next action. - Delivery verification — confirms 23 roles, gates, and routes; validates all checkpoint evidence and handoff hashes; surfaces unknown, unavailable, failed, or stale stages and known limitations before the user decides to accept, hold, or rework.
How to use it
Start by running python scripts/lifecycle.py intake --root <project-root> to classify your project's state. Invoke Light Orchestrator when you begin a new research project, resume work, checkpoint a stage, encounter a failed gate, or recover from staleness. Hand it the project root, the passport file path, and producer findings (as JSON); the skill will guide you through state decisions, stage routing, reroute authorization, and delivery verification without executing tasks or imposing research choices.
Related skills
Documentation Co-Authoring
anthropics
Guide structured workflows for writing docs, proposals, and technical specs collaboratively.
MCP Server Builder
anthropics
Build protocol servers that connect language models to external APIs and services.
Ask Questions If Underspecified
trailofbits
Ask clarifying questions before starting work on ambiguous requests.
Claude Plugin Marketplace Builder
daymade
Convert a Claude skills repository into an installable marketplace with auto-updates.