菜单
Run live end-to-end verification across Morphic-Agent's 6 agent CLI engines. Produces a baseline comparison table with duration and cost per engine.
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。
Verify a feature spec, plan, or PR against the non-negotiable principles in `.specify/memory/constitution.md`.
Extract session insights after a task completes and feed them into Morphic's self-evolution engine. Captures what worked, what didn't, and proposes prompt/routing tweaks.
Keep CLAUDE.md (thin router) and AGENTS.md (telegraph-style) in sync. Detects drift in command lists, paths, and routing tables between the two root instruction files.
Generate a monthly or on-demand cost report across LLM providers, local LLMs, and agent engines. Flags overspend and suggests downgrades.
Inspect Morphic-Agent's fractal task graph for a given task_id. Reports bypass classifier decisions, Gate 2 skips, parallelism, and regression flags.
Preview a LAEE action plan — classify risk, check undo availability, inspect credential touchpoints — without executing anything.
基于 SOC 职业分类
| name | engine-e2e |
| description | Run live end-to-end verification across Morphic-Agent's 6 agent CLI engines. Produces a baseline comparison table with duration and cost per engine. |
| when_to_use | When the user asks for an "E2E round", "live test", "engine check", or "smoke test". |
| argument-hint | [engines | all] |
| allowed-tools | ["Bash","Read","Grep"] |
| model | sonnet |
Runs the canonical smoke task against each available engine and produces a comparison report.
/engine-e2e # all engines
/engine-e2e ollama # single engine
/engine-e2e claude,codex,gemini # multiple
Argument: $ARGUMENTS
"Return the sum 2+2 as a single number."
Expected output: exactly 4 (or 4. / "4" depending on engine format).
Delegate to the engine-tester subagent with the engine list from $ARGUMENTS.
The subagent runs each engine in a forked context (keeps noise out of main log).
Collect its report.
Append a one-line entry to docs/CONTINUATION.md under "Live Verification Summary":
- **Round NN** (<date>): <description>, N/M engines passing, $X.XX
If any engine regressed vs the last known-good baseline, flag it.
| Engine | Duration | Cost | Source |
|---|---|---|---|
| Ollama (qwen3:8b) | ~1-3s | $0.00 | Round 17 |
| Claude Code SDK | ~3-5s | ~$0.01 | Round 15 (via routing) |
| Codex CLI | ~40-50s | ~$0.02 | Round 15 |
| Gemini CLI | ~2-5s | ~$0.001 | Round 18 |
| OpenHands (Gemini) | ~30-60s | ~$0.005 | Round 16 |
| Fractal (hybrid) | ~5-10s | ~$0.024 | Round 17 |
Table + summary + delta vs baseline. Store in docs/CONTINUATION.md.