Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Secureclaw Scanner
v1.0.4Native security prompts and best practices to instantly make your OpenClaw instance safer without relying on 3rd party APIs.
⭐ 0· 89·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name/description (security prompts & best practices) matches the package contents: prompt files and markdown checklists. There are no unexpected env vars, binaries, or install actions requested.
Instruction Scope
SKILL.md and PROMPTS.md stay within scope: they recommend manual reviews, refuse dangerous commands, and advise checking ~/.openclaw/openclaw.json. They do reference that path but explicitly instruct the agent not to read/modify it automatically, which is proportionate.
Install Mechanism
No install spec or code to fetch/execute; instruction-only skill means nothing is written to disk by an installer. This is the lowest-risk install mechanism.
Credentials
The skill requires no environment variables, credentials, or config paths beyond recommending the user manually inspect their OpenClaw config. No broad secret access is requested.
Persistence & Privilege
always is false and autonomous invocation is default. The skill intends to be included in runtime context window only, which is normal for prompt-based guardrails and does not demand elevated persistence or cross-skill config changes.
Scan Findings in Context
[ignore-previous-instructions] expected: The regex scanner flagged the 'ignore previous instructions' pattern. In context the skill explicitly calls out that phrase to instruct agents to reject such prompt-injection attempts, so its presence is expected and appropriate.
Assessment
This skill is internally coherent and does not request secrets or install code, but remember: it only provides guidance inside the model's context window — it cannot enforce OS-level sandboxing. Before installing: (1) verify the package source (homepage/ClawHub listing) and publisher identity, (2) keep terminal/network tools disabled or restricted unless absolutely required, (3) manually inspect ~/.openclaw/openclaw.json and maintain API authentication and localhost binding, and (4) treat these prompts as a helpful layer, not a replacement for containerization, human approval workflows, or external security tooling. If you need enforcement (not just advice), use platform controls or external monitor/agent hardening tools.PROMPTS.md:9
Prompt-injection style instruction pattern detected.
SKILL.md:40
Prompt-injection style instruction pattern detected.
About static analysis
These patterns were detected by automated regex scanning. They may be normal for skills that integrate with external APIs. Check the VirusTotal and OpenClaw results above for context-aware analysis.Like a lobster shell, security has layers — review code before you run it.
latestvk977z652p1wpk46ek3s9myv5n983xkc4
License
MIT-0
Free to use, modify, and redistribute. No attribution required.