菜单
Use when the user says "loop", "반복", "쭉 돌려줘", or wants to run triage then resolve repeatedly until there are no agent-workable llmux issues/PRs left. Thin orchestrator over the agent-triage and agent-resolve skills; dry-run by default.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when the user says "resolve", "해결해줘", "픽스해줘", or wants to take one ready-to-agent llmux issue/PR, implement the fix on its branch, get `just check` green, and hand back a review-ready PR. Does NOT merge or deploy by default.
Use when the user says "triage", "이슈 정리", "분류", or wants to scan open llmux GitHub issues, decide which are safe for an agent to implement, label them, and open a tracking branch + draft PR for each ready one. Reads and labels issues; never writes source code.
Use for normal llmux dev iteration when the user says "빌드", "빌드해줘", "build", "build and deploy locally", or wants to compile + run a change locally and save it. Builds release, hot-deploys the binary to the local daemon, commits, and pushes to a feature branch (never main).
Use when the user says "배포", "배포해줘", "deploy", or "ship a preview" for llmux. Pushes to main, lets CI publish a preview prerelease, refreshes the llmux-preview brew formula, verifies brew actually updated, then hot-deploys that build locally and restarts.
Use when the user says "릴리즈", "릴리즈해줘", "release", "cut a release", or "stable release" for llmux. Bumps the version, tags v*, lets CI publish the stable GitHub release, refreshes the llmux stable brew formula, verifies brew updated, hot-deploys + restarts locally, and verifies client + server with llmux status.
| name | agent-loop |
| description | Use when the user says "loop", "반복", "쭉 돌려줘", or wants to run triage then resolve repeatedly until there are no agent-workable llmux issues/PRs left. Thin orchestrator over the agent-triage and agent-resolve skills; dry-run by default. |
agent-loop has no judgment of its own. It runs agent-triage once, then agent-resolve on the
resulting ready PRs one at a time, and stops on clear conditions. Labels are its memory — it
never re-evaluates issues already marked needs-* / not-in-repo / agent-blocked. Inherits
agent-resolve's boundary: it does not merge or 배포.
--apply.agent-triage once. (Labels issues, opens draft PRs for ready ones, skips already-labeled.)ready-to-agent draft PRs, oldest first, run agent-resolve — each in its own
git worktree (per that skill), so resolves are isolated and may run in parallel. Verify
each PR genuinely green (just check + CI) independently before counting it done. Parallelize
only across issues that touch disjoint modules; if their diffs would overlap, serialize them to
avoid merge conflicts.ready-to-agent draft PRs remain (the user's "no issues/PRs left" condition),--max-iterations reached (default 3),agent-resolve runs end in agent-blocked,agent-resolve run hit a kill-switch needing a human (needs-design / needs-human / security / release).deploy).Never re-triage an issue that already carries a needs-* / not-in-repo / agent-blocked label —
a human must clear the label for it to re-enter the pipeline. Without this, the same un-fixable issue
(e.g. the gpt-5.5 client-limitation one) gets re-evaluated to the same dead end every iteration.
agent-loop inherits agent-resolve's human-gated boundary.--max-iterations and the stop conditions.