Longrun — AI skill for Claude Code
/longrun — drain the feature queue with model-routed workers.
How to install Longrun
Installs to ~/.claude/skills/fj-osprey-claude-personal-os-longrun/SKILL.md
mkdir -p ~/.claude/skills/fj-osprey-claude-personal-os-longrun && curl -fsSL https://raw.githubusercontent.com/FJ-Osprey/claude-personal-os/HEAD/commands/longrun.md -o ~/.claude/skills/fj-osprey-claude-personal-os-longrun/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What Longrun does
/longrun — drain the feature queue with model-routed workers
Geoff said `longrun` (or its dictation alias `runqueue` / "run queue"). This is the **only** trigger that starts an autonomous queue-draining run. Everything about the system: `docs/MODEL-COORDINATION.md` in the project (games has it today).
Modes — read the argument text
**This table is canonical.** `~/.claude/CLAUDE.md` and `docs/MODEL-COORDINATION.md` point here and must not restate it; if you change a mode, change it only h
Alternatives in AI
- Cursor Tools — by eastlondoner - Creates a versatile AI command interface supporting multiple providers and models with flexi 4.7k ★
- Launch An AI Feature — Workflow recipe — take an AI/LLM feature from a probabilistic-aware PRD to a launch-ready model card by chaini 1.3k ★
- Agentlas Orch — Set or show which model runs the orchestrator and which runs the workers 1.1k ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Multi Agent Orchestrator
Multi-model orchestrator: runs Claude Code, Codex, Gemini, DeepSeek, GPT, Qwen and more as interchangeable wor
Local Drain
Run the autonomous LLM-orchestrator restricted to LOCALLY-EXECUTABLE plans — the two oracle lanes, both drain-
Skill Router
Skill + Agent + Model + Thinking depth — auto-routed before any tool fires. One SKILL.md for Claude Code. 90%
Route Feedback
Tell decision-router which model the last routed task should have used
Jev Linkmap
Rebuild a site's internal link map in seconds with Jev, race Claude Opus 5 on the same queue, and let a deep m
Fleet Wait
Wait for a run's workers to finish without spending model turns, then collect. Use while workers are running,
Related Agents
GPT 5.6 Sol
General-purpose agent that runs on GPT-5.6-Sol (routed through the shunt gateway to the ChatGPT/Codex subscrip
Routed Impl
Implementation-class dispatch target — production code, multi-file integration, refactors. Functionally equiva
GPT 6 Luna
General-purpose agent that runs on GPT-6-Luna (routed through the shunt gateway to the ChatGPT/Codex subscript