一键导入
leopold-stop
Take the seat back. Cleanly ends the autonomous run at the next turn boundary and writes a final summary. Nothing is interrupted mid-turn.
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。
菜单
Take the seat back. Cleanly ends the autonomous run at the next turn boundary and writes a final summary. Nothing is interrupted mid-turn.
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
Phase 1 of Leopold. A structured debate that captures the mission and your decision-making, then writes the brief (MISSION, CHARTER, GUARDRAILS, PLAN) that the autonomous run executes.
Launch a local live web dashboard to watch the current project's Leopold run — status, cost meters (context/subagents/forks), event feed, decisions, and a Stop button. Local-only (127.0.0.1), zero dependencies.
Compile a Leopold brief into a dynamic workflow and run it. The plan lives in code instead of one growing context window, every item gets an independent adversarial review, and git stays locked — nothing is committed. The workflow is resumable and shows up in /workflows.
Control plane for the global prompt enhancer: status, on/off, preview a prompt against the gate, and the learn loop that mines your enhancement ledger into proposed prompt-profile rules. Never edits PROMPT-PROFILE.md itself — you review and apply.
The self-improving charter. Mines past Leopold decisions, session corrections, and git history for recurring judgment calls, skeptic-verifies each candidate, and proposes charter amendments. Never edits CHARTER.md itself — you review and apply.
Phase 2 of Leopold. Enters autonomous mode and conducts Claude Code through the plan, deciding from the charter instead of asking, with git locked. The Stop hook keeps it going until the plan is done or a stop condition fires.
| name | leopold-stop |
| version | 0.1.0 |
| description | Take the seat back. Cleanly ends the autonomous run at the next turn boundary and writes a final summary. Nothing is interrupted mid-turn. |
| allowed-tools | ["Read","Write","Bash"] |
| triggers | ["leopold stop","stop leopold","take back the seat"] |
End the autonomous run cleanly.
Run:
LEO=.leopold
if [ ! -f "$LEO/state.json" ]; then echo "No Leopold run to stop."; exit 0; fi
tmp="$(mktemp)"; jq '.active=false | .stopped_reason="user_stop"' "$LEO/state.json" > "$tmp" && mv "$tmp" "$LEO/state.json"
printf '{"ts":"%s","event":"stop","reason":"user_stop"}\n' "$(date -u +%Y-%m-%dT%H:%M:%SZ)" >> "$LEO/events.jsonl"
rm -f "$LEO/STOP" "$LEO/ALLOW_GIT" "$LEO/ALLOW_PUSH"
echo "Leopold run stopped (git re-locked; STOP and opt-in tokens cleared)."
This sets the run inactive, so the Stop hook will allow the session to halt at
the next turn boundary. The blunt alternative is touch .leopold/STOP.
After stopping, give the user a short handoff: what was completed (done items in
PLAN.md), what decisions were made (DECISIONS.md), and what is staged and
ready for them to review and commit. Remember: Leopold never committed; that is
their call.