jleechanorg
jleechanorg/claude-commands
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 openclaw-models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 | openclaw-models |
| description | OpenClaw agent model configs — which work, which are broken/quota-limited, and how to switch |
| type | reference |
Live config: ~/.openclaw/openclaw.json → agents.defaults.model
Auth profiles: ~/.openclaw/agents/main/agent/auth-profiles.json
"model": {
"primary": "minimax/MiniMax-M2.7",
"fallbacks": []
}
timeoutSeconds: 900 (M2.7 regular can be slow; 600 was too low)
| Model | Status | Auth type | Notes |
|---|---|---|---|
minimax/MiniMax-M2.7 | ✅ WORKING — current primary | api_key → minimax:default | Can be slow; timeout set to 900s |
minimax/MiniMax-M2.7-highspeed | ❌ PLAN NOT SUPPORTED | api_key → minimax:default | HTTP 500 error 2061 — current API key plan does not include this model |
openai-codex/gpt-5.3-codex | ❌ QUOTA-LIMITED | OAuth → openai-codex:default | Weekly usage cap exhausts; DO NOT use as primary/fallback |
openai-codex/gpt-5.3-codex-spark | ⚠️ Same quota | OAuth → openai-codex:default | Same weekly pool as gpt-5.3-codex; used by consensus agent |
xai/grok-4-fast | ❓ UNVERIFIED | api_key → XAI_API_KEY env | Key was flagged 403/revoked 2026-03-28; verify before using |
xai/grok-3-mini | ❓ UNVERIFIED | api_key → XAI_API_KEY env | Same key as above |
openrouter/auto | ❓ NOT CONFIGURED | api_key → OPENROUTER_API_KEY | Key not in openclaw.json env; add before using |
anthropic-vertex/claude-sonnet-4-6 | ❓ NOT CONFIGURED | GCP credentials | Needs gcp-vertex-credentials profile; not set up |
"minimax:default": {
"type": "api_key",
"provider": "minimax",
"key": "[REDACTED_OPENAI_KEY]"
}
Critical: must be "type": "api_key" (underscore) and "key" (not "apiKey").
Before writing to openclaw.json, verify the model is available on the current plan:
# Get the current MiniMax API key from auth-profiles
KEY=$(python3 -c "import json; d=json.load(open('$HOME/.openclaw/agents/main/agent/auth-profiles.json')); print(d['profiles']['minimax:default']['key'])")
# Probe the candidate model — expect HTTP 200, not 500
curl -s -o /dev/null -w "%{http_code}" \
https://api.minimax.io/anthropic/v1/messages \
-H "x-api-key: $KEY" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{"model":"MiniMax-M2.7-highspeed","max_tokens":5,"messages":[{"role":"user","content":"hi"}]}'
Never add a model to openclaw.json without a 200 probe response.
import json
path = "$HOME/.openclaw/openclaw.json"
with open(path) as f: d = json.load(f)
d['agents']['defaults']['model']['primary'] = "minimax/MiniMax-M2.7"
d['agents']['defaults']['model']['fallbacks'] = []
with open(path, 'w') as f: json.dump(d, f, indent=2)
kill -9 $(lsof -ti :18789); sleep 12; lsof -i :18789 | grep LISTEN
grep "agent model" /tmp/openclaw/openclaw-$(date +%F).log | tail -2
agents.defaults.timeoutSeconds: 900 — MiniMax M2.7 (regular) can be slow; 600s was insufficient.
Do not lower below 900s while on M2.7 regular.
| Symptom | Cause | Fix |
|---|---|---|
| :eyes: reaction but no reply | MiniMax M2.7 timeout exceeded | Increase timeoutSeconds (currently 900); check logs for FailoverError |
HTTP 500 error 2061 in logs | Model not on current API key plan | Mark ❌ in status table; do NOT add to config; probe before switching |
LiveSessionModelSwitchError | Model changed while session was live | Expected after restart; clears on next run |
FailoverError: LLM request timed out | Primary timed out, no working fallback | Highspeed ❌ on this plan; investigate xAI grok or OpenRouter |
Profile minimax:default timed out. Trying next account... | MiniMax slow; fallback also failed | Check fallback list — remove unsupported models |
Codex weekly usage exhausted | ChatGPT Pro plan weekly cap | Switch to minimax immediately |