菜单
General read-only execution tracer. Profiles automate known systems; without a profile the LLM must fresh-investigate entry→hops→blocker from live evidence. Triggers on /callpath, trace execution, where stuck.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
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。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/jleechanorg/claude-commands --skill callpath命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Token-efficient second opinion slash command /advice. Extracts decision point + artifact (≤150 lines), then fans out in parallel: (1) Opus subagent reviewer with fallback chain codex→agy→cursor, (2) /research on the decision topic, (3) /secondo multi-model opinion. Use instead of advisor() which ships the full conversation uncached.
Use this skill when working in repositories managed by Agent Orchestrator or when the user asks how to use `ao` properly. Covers the default AO workflow: bootstrap with `ao start`, dispatch work with `ao spawn`, inspect progress with `ao status` or `ao session ls`, steer sessions with `ao send`, and recover or clean up sessions safely. Includes strict parameter fidelity, pre-spawn cap cleanup, quota-wall fallback, and post-spawn verification.
Generate a full agento PR status report — draft readiness, canonical /green, zero-touch rate, inline display, and Slack summary.
基于 SOC 职业分类
正在显示 SKILL.md
| name | callpath |
| description | General read-only execution tracer. Profiles automate known systems; without a profile the LLM must fresh-investigate entry→hops→blocker from live evidence. Triggers on /callpath, trace execution, where stuck. |
Trace where work entered, which components handled it, and where it stalled — any repo or multi-service stack.
callpath list
callpath run <profile> [args...]
Profiles automate recurring multi-hop checks (dark-factory, future hermes-request, etc.). Run the profile, interpret stdout, extend only if gaps remain.
Do not skip to callpath trace with guessed hops. Do not reuse prior chat conclusions, memory, or “usual” paths without re-verifying in this session.
The agent must investigate from scratch:
trace_id, bead id, webhook, cron, user action). Read configs, code paths, recent logs, gh, ao, health endpoints as needed.Optional: after investigating, document probes with:
callpath trace --title "..." --hop 'id:component:cmd' ...
That command records hops you already validated — it does not replace investigation.
entry → hop₁ → hop₂ → … → outcome — see schema/trace-model.md.
profiles/<name>/profile.yaml + run.sh. Copy profiles/_template/ for new systems.
dark-factory (one profile among many):
callpath run dark-factory --prs 8058,8060
# /callpath — <timestamp> — verdict=<GREEN|AMBER|RED>
entry: <correlation id + how you found it>
## Hops
<id> <status> [<component>] <one-line fresh evidence>
blocker: <first FAIL hop or "none">
next: <one concrete read-only or operator step>
/callpath