YES.md 日本語版
v1.1.0ファイル・設定・データベース・デプロイの変更を伴うタスクで発動。デバッグが2回以上連続で失敗した時に発動。証拠なしに推測・仮定しようとした時に発動（「おそらく」「多分」「〜だと思う」「〜のはず」）。ユーザーに丸投げしようとした時に発動（「ご確認ください」「手動で対応してください」「〜が必要かもしれません」）。修正...
⭐ 0· 209·0 current·0 all-time
bytkman@sstklen
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The skill is a rules-and-checklist engine for safe debugging, edits, and deployments; the commands and checks it describes (cp for backups, grep/lsof/env/node -v/curl for verification, WebSearch for research) are appropriate and proportional to that purpose.
Instruction Scope
SKILL.md instructs the agent to run shell commands and read configuration/log files and to use WebSearch. This is expected for a change-safety governance skill, but it assumes the agent has filesystem and network tooling available — if the agent environment lacks or restricts these capabilities, the skill cannot fully operate.
Install Mechanism
No install spec and no code files — the skill is instruction-only, which minimizes installation risk (nothing is downloaded or written to disk by the skill itself).
Credentials
The skill declares no required environment variables, credentials, or config paths. While it suggests using env/cat/grep/etc. to gather evidence, it does not demand secrets or unrelated credentials, so requested access is proportional to its stated purpose.
Persistence & Privilege
always is false and the skill does not request persistent system-wide privileges or modify other skills; autonomous invocation is allowed by default but not elevated here.
Assessment
This skill is a conservative, evidence-first checklist for debugging and making safe changes and appears internally consistent. It will expect the agent to be able to run shell commands, read files, and perform WebSearches — capabilities that can expose sensitive local data if the agent is allowed full filesystem or network access. If you plan to enable it, ensure the agent runs in an environment where those actions are appropriate (sandboxed or limited access) and that you trust the agent's execution context; otherwise the skill is safe but may be unable to perform all checks.Like a lobster shell, security has layers — review code before you run it.
debuggingvk971yghww9vcc7p56c1gdneqd182pm8egovernancevk971yghww9vcc7p56c1gdneqd182pm8ehooksvk971yghww9vcc7p56c1gdneqd182pm8ejapanesevk971yghww9vcc7p56c1gdneqd182pm8elatestvk971yghww9vcc7p56c1gdneqd182pm8emultilingualvk971yghww9vcc7p56c1gdneqd182pm8esafetyvk971yghww9vcc7p56c1gdneqd182pm8e
License
MIT-0
Free to use, modify, and redistribute. No attribution required.