Security audit

Clawhub

Security checks for vulnerabilities and agentic risk

Overview

The scanners classify this skill as benign because its only reported issue concerns publishing scope, while the other verdict dimensions are rated ok; confidence is medium and coverage is partial.

The verdict is benign, but review what would be included and where it would be published before using bulk synchronization. Confidence is medium and the scan was partial: two local path-like references could not be resolved, so the evidence is incomplete.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

The publishing workflow instructs bulk synchronization without warning that local skill contents may be uploaded to the public registry.

Medium

Category: Output Handling
Confidence: 90% confidence
Finding: The publishing workflow presents clawhub sync --all without warning that bulk synchronization may upload local skill contents to the public registry or requiring confirmation of the upload scope. This could disclose private instructions, proprietary material, or credentials embedded in eligible skill files. Its placement in a legitimate publishing section reduces suspicion of malicious intent, but does not establish authorization to publish every local skill.
Content: No source excerpt is available for this finding.