一键导入
ao-agent-shorthands
Read ~/.hermes/agent-orchestrator.yaml BEFORE declaring an --agent <X> value unsupported. Lists plugin shorthands (wafer, minimax, agy).
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
Read ~/.hermes/agent-orchestrator.yaml BEFORE declaring an --agent <X> value unsupported. Lists plugin shorthands (wafer, minimax, agy).
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。
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.
Lazy senior-dev mode — the seven-rung ladder that decides whether to write code at all, whether to reuse code that already exists, and whether to prefer stdlib/platform/installed deps over a new dependency. Use before writing any code, every PR diff, every fix, every feature. Source attribution included.
Use when making any "X is enabled" / "feature X works" / "cache works" claim in a running session — requires runtime probe output from the standard harness startup path, NOT a launchd/cron test that explicitly sets the activation env var.
Pre-flight checklist for self-hosted runner failures. Use when investigating low disk, missing runner, or stuck Green Gate on the jleechanorg/worldarchitect.ai fleet. Always verifies host-level container state, not just GitHub API.
When asked to verify whether a Claude Code feature works (especially slash commands, dialogs, pickers, status indicators), spawn a real interactive TUI session in cmux — never use `claude --print "/feature"` as a test, because --print is non-interactive and will always return "isn't available in this environment" regardless of whether the feature actually works.
Use whenever running an AO spawn. Honors exact --agent/--runtime/--project/--claim-pr; verifies session metadata; never silently substitutes a different worker.
| name | ao-agent-shorthands |
| description | Read ~/.hermes/agent-orchestrator.yaml BEFORE declaring an --agent <X> value unsupported. Lists plugin shorthands (wafer, minimax, agy). |
MANDATORY — resolve --agent from config before declaring unsupported:
When the user specifies --agent <X> and <X> is not a well-known CLI name you recognize, read ~/.hermes/agent-orchestrator.yaml (or the active AO config) FIRST. The defaults.agent key and installed plugins define agent shorthands. Never scan packages/agent-*/ directories as a substitute. Never declare an agent unsupported without reading the config.
# Step 1 — always do this when --agent value is unfamiliar:
head -20 ~/.hermes/agent-orchestrator.yaml # check defaults.agent
| Shorthand | Binary / mechanism | Notes |
|---|---|---|
--agent agy / --agent antigravity | ~/.local/bin/agy (Google Antigravity / Gemini CLI) | Default agent in ~/.hermes/agent-orchestrator.yaml (defaults.agent: antigravity) |
--agent wafer | claude CLI → https://pass.wafer.ai | WAFER_API_KEY, default model GLM-5.1 |
--agent minimax | claude CLI → https://api.minimax.io/anthropic | MINIMAX_API_KEY, server-selected model |
z.ai/model prefix | claude → https://api.z.ai/api/anthropic | GLM_API_KEY, model in prefix |
The claude-code plugin supports inline model prefixes for all three: wafer.ai/GLM-5.1, z.ai/claude-4, MiniMax-M2.7. Dedicated --agent plugins are preferred for wafer and minimax. Z.AI uses inline prefix only (no dedicated agent plugin).
Integration tests: packages/integration-tests/src/agent-{wafer,minimax,zai}.integration.test.ts — real tmux + real API, auto-skip when keys missing.