Use when updating OpenClaw or debugging an OpenClaw instance after an update. This skill acts as a structured update runbook with emphasis on gateway startup...
21 次安装2 星标
编程
Openclaw Skill Auto Update Discovery
试用Engine auto-update & discovery skill OpenClaw: scan, evaluate, update, test, install, verify, rollback — dengan validation gate dan safety agar update tidak merusak workspace.
它能做什么
Engine auto-update & discovery skill OpenClaw: scan, evaluate, update, test, install, verify, rollback — dengan validation gate dan safety agar update tidak merusak workspace.
技能文档
Edge Cases
- ClawHub CLI tidak tersedia → fallback
git+web_fetch; laporkan tool hilang, jangan gagal diam. - Workspace di Termux vs desktop → path & binary beda; jangan copy skill bawaan (
npm-global). - Skill punya dependency tool yang tak ada → BLOCK, jangan install parsial.
- Dua skill dengan capability identik → ranking by Quality Score + Relevance, pilih 1, disable duplikat (bukan hapus).
- Update mayor dgn migration → PENDING REVIEW, jangan auto-apply.
- Rollback tapi backup rusak → ESCALATE ke user, jangan asumsi versi aman.
Concrete Examples
- Input: "Cek update buat trading-analyst." → Output: scan
_meta.json(v1.0.0),clawhublatest (v1.1.0), baca changelog (fix RSI), security scan PASS, compat PASS → lapor "update v1.0.0→v1.1.0 tersedia, low risk" → tunggu approval / auto-update bila gate lolos → backup → apply → post-install test (trigger "analisa gold" respon) → VERIFY PASS → report. - Input: "Cari skill buat screenshot web di Android." → Output: discover by capability "screenshot/web/android" → temukan 2 kandidat → Quality Score (A=82, B=54) → eval B (obfuscated install script) → BLOCK B → rekomendasikan A, minta approval install.
- Input: "Skill brain-core-ultra error load." → Output: detect broken → diagnose (JSON corrupt di
_meta.json) → cek latest source → patch/copy bersih → post-install test → jika gagal → ROLLBACK ke backup → report.
Failure Modes
- Auto-update buta → break agent. FIX: validasi gate wajib sebelum apply.
- Version conflict tak diresolve → skill saling timpa. FIX: prioritas + PENDING REVIEW.
- Tanpa rollback → rusak permanen. FIX: backup old version tiap update penting.
- Update saat agent aktif pakai skill → race condition. FIX: jangan update skill sedang dipakai, atau queue.
- Hallucinate "installed" → laporan bohong. FIX: anti-hallucination, verify dulu.
- Bloat (10 skill sama) → resource habis. FIX: anti-bloat, pilih 1 terkuat.
Common Mistakes
| Mistake | Fix |
|---|---|
| Auto-updating without validation | Validate before applying |
| Version conflicts | Resolve priorities |
| No rollback | Keep previous versions |
| Updating broken skills | Detect and skip broken updates |
Red Flags
- Applying untested updates
- Ignoring version conflicts
- No rollback capability
- Updating during active use
Rationalization Prevention
| Excuse | Reality |
|---|---|
| "Updates are safe" | Validate each. |
| "Latest wins" | Resolve conflicts. |
| "I'll rollback if needed" | Have rollback ready. |
How to Use
- Scan: Discover skills and available updates.
- Evaluate: Assess version conflicts and compatibility.
- Update: Apply with validation gates.
- Verify: Test post-update; rollback if broken.
Quick Reference
| Situasi | Aksi |
|---|---|
| Cek update skill | Jalankan discovery |
| Update tersedia | Validasi sebelum apply |
| Versi bentrok | Resolve prioritas |
| Auto-update gagal | Rollback aman |
| Selesai update | Verifikasi berfungsi |
相关技能
Install or upgrade a ClawHub skill, then make the global copy under the machine's OpenClaw home `skills/` directory the final source of truth. Use when the u...
14 次安装
Explain how to use OpenClaw in a local environment. Use when the user asks how to find OpenClaw config or workspace files, how to add or inspect skills, how...
16 次安装
Ship public OpenClaw docs and skill files with fewer leaks, fewer empty claims, and clearer operator instructions.
6 次安装
Record a screen demonstration and turn it into a reusable, parameterized OpenClaw SKILL.md.
The master orchestrator for the OpenClaw Skill OS ecosystem. Coordinates multiple skills, manages skill interactions, and routes tasks to optimal skill combinations when the user explicitly asks to orchestrate or combine skills. Use this skill only when the user requests skill orchestration, multi-skill routing, or ecosystem-wide coordination — not for ordinary single-skill tasks.