ClawHub
Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

Navil Audit

v1.0.2

Deep security audit for OpenClaw. Run a comprehensive scan of all installed skills, MCP servers, and agent configuration. Generates a detailed security repor...

⭐ 0· 47·0 current·0 all-time
by@ivanpantheon
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Suspicious
medium confidence
✓
Purpose & Capability
Name, description, and runtime instructions consistently describe a local security scanner and pentest toolkit (navil). Requiring pip to install a CLI scanner is reasonable and proportionate to the stated purpose.
!
Instruction Scope
Instructions ask the agent to 'scan all installed skills, MCP servers, and agent configuration' but do not require the user to supply explicit paths or consent; placeholders like <path-to-config> are left ambiguous. The skill delegates network queries (VirusTotal, pentest simulations) to the navil CLI without documenting network behavior or data transmission—this ambiguity could lead to sensitive config data being sent externally by the installed tool.
ℹ
Install Mechanism
SKILL.md contains a pip install spec for a package named 'navil' (bins: navil), which is a common and expected install method for a Python CLI. However, the registry metadata reported 'No install spec — instruction-only', creating an inconsistency between package metadata and SKILL.md. Installing a PyPI package is higher-risk than an instruction-only skill because it executes third-party code; verify the package source (GitHub link present) before installing.
✓
Credentials
The skill does not request any environment variables, credentials, or config paths in metadata. This is proportionate. Note: some checks (VirusTotal, remote pentest checks, MCP server scanning) may implicitly require credentials or network access at runtime; none are declared.
✓
Persistence & Privilege
always:false and normal model invocation are used. The skill does not request persistent presence or system-wide configuration changes. It does instruct scanning of local skill configs and MCP servers, which requires file access but not elevated privilege by default.
What to consider before installing
What to consider before installing/running Navil Audit: - Verify the navil pip package: review the linked GitHub repo (https://github.com/navilai/navil) and confirm the PyPI package name and publisher are legitimate before pip installing. - Run the tool in a sandbox or isolated environment first (container/VM) because pip-installed CLIs run arbitrary code and may perform network calls. - Clarify scan targets: do not let the tool search your entire filesystem automatically. Provide explicit config paths and confirm which MCP servers and skill directories will be scanned. - Expect that the navil CLI may query external services (VirusTotal, vulnerability feeds). If you have sensitive configs, ensure network access is controlled or redact secrets before scanning. - Check for the metadata inconsistency: SKILL.md contains an install block while registry metadata lists no install spec—ask the publisher to reconcile this. - For CI use, validate SARIF output handling to avoid leaking secret data into logs or external scanners. If you want a safer evaluation, request the navil package source code (or a pinned release tarball) and have it reviewed before allowing the agent to install or run it.

Like a lobster shell, security has layers — review code before you run it.

latestvk976g2xyfy8kk55fdvtdk2kf7583k6tgmcpvk9704crwzhvhrmn7hjmxmyxrxn83jyagsecurityvk9704crwzhvhrmn7hjmxmyxrxn83jyag

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Runtime requirements

🔍 Clawdis
Binspip

Comments