Skill Install Guardian

AdvisoryAudited by Static analysis on May 10, 2026.

Overview

Detected: suspicious.dynamic_code_execution

Findings (1)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

NoteHigh Confidence

ASI02: Tool Misuse and Exploitation

What this means

Running the skill will execute local `npx clawhub` inspection commands and fetch information about the requested skill.

Why it was flagged

The script invokes a local CLI command to inspect ClawHub skills. This is expected for the scanner purpose, and the visible code validates the slug and uses list-form subprocess arguments.

Skill content

stdout, stderr, code = run_command(["npx", "clawhub", "inspect", slug, "--json"])

Recommendation

Use it only for intended skill slugs, review the generated report, and keep the owner-confirmation step before any installation.

NoteMedium Confidence

ASI04: Agentic Supply Chain Vulnerabilities

What this means

If `npx` resolves or updates the CLI from an external package source, the scanner depends on that package supply chain.

Why it was flagged

The skill relies on `npx` to run the ClawHub CLI, but the artifact does not pin a CLI package version or source. This is common and purpose-aligned, but it means the trust boundary includes the local or resolved `clawhub` CLI.

Skill content

requires":{"bins":["npx"]} ... npx clawhub inspect <skill-slug> --security

Recommendation

Prefer a trusted, pinned, or preinstalled ClawHub CLI when using this as a security gate.

Findings (1)

critical

suspicious.dynamic_code_execution

Location: scripts/check.py:17
Finding: Dynamic code execution detected.