Security audit
Security checks for vulnerabilities and agentic risk
This skill is a local deployment record helper that writes service notes but does not request credentials, install code, or transmit data externally.
Install this if you want agents to maintain a local OpenClaw service registry. Be aware it may write deployment metadata such as ports, URLs, dependency names, and credential locations into memory/services/, so avoid using it in projects where that operational inventory should not be persisted.
Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
✅ 正确：
- `token` 环境变量，配置在 `.env` 文件
- `AUTH_CODE` 启动参数，见 docker-compose.yml
- 凭证路径：`~/.config/xxx/credentials.json`（字段名不暴露）
## 示例模板
The description says to use the skill whenever deployment is added, services are queried, statuses are updated, or dependencies are checked, and the body repeats that it triggers in those cases. These are broad operational situations without explicit constraints, exclusion conditions, or negative examples, which could cause unintended invocation during ordinary discussions about services.
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
curl http://localhost:4001/api/healthThe natural-language description on L12 is written only in Chinese, which can impose a language/locale constraint without offering user opt-in or documenting a justified region-specific scope. The policy requires flagging language or locale restrictions unless the skill explicitly offers choice or clearly documents why the locale constraint is necessary.
This markdown file is entirely written as a prescriptive documentation standard in Chinese, including required field names and examples, but it does not indicate that language is optional or limited to a justified Chinese-only context. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation.
No suspicious patterns detected.