skill-shield

clawhub:skill-shield

View source
C
55/100

First Seen

Feb 20, 2026

Last Scanned

Feb 22, 2026

Findings

3

Score

55/100

HIGH 3

Findings (3)

HIGH
Private data read with code execution
L51

Skill can read private data AND execute arbitrary code. This combination enables credential theft via dynamic code.

[reads_private_data] /etc/passwd + [executes_code] exec(
FIX

Add input validation between the user-controlled data source and the security-sensitive sink (e.g., file writes, command execution). Implement allowlisting for acceptable input patterns.

FP?

Likely FP if the user input passes through explicit validation or sanitization before reaching the sensitive operation, and the taint tracker missed the sanitization step.

HIGH
DNS exfiltration pattern
L46

Detects DNS-based data exfiltration techniques

DNS exfil
FIX

Limit clipboard access to write-only operations or require explicit user confirmation before reading clipboard contents. Block transmission of clipboard data to external endpoints.

FP?

Likely FP if the skill is a clipboard manager whose documented purpose is to sync clipboard content with the user's own devices.

HIGH
Privilege escalation
L49

Detects privilege escalation patterns like setuid, chown root, or sudo with shell commands

setuid
FIX

Avoid depending on packages that could be subject to typosquatting or name confusion. Verify package ownership, check download counts, and audit the package source before adding dependencies.

FP?

Likely FP if the flagged package is a well-known, high-download-count package from a verified publisher.