ClawHub
Back to skill

Security audit

insight

Security checks for vulnerabilities and agentic risk

Overview

This is a straightforward note-analysis skill, but users should be aware it can read sensitive notes and may write analysis back into note files by default.

Before installing, assume the skill may process private journals, relationship notes, and other sensitive content. Use `reply-only`, narrow `--source`, `--dir`, `--time`, and `--count` settings when possible, and avoid feeding crisis, medical, or third-party private information unless you have a clear reason and consent.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The README explicitly promotes analysis of personal notes and includes mental-health-adjacent outputs such as 'Thought Trap Audit,' 'Blind Spot Reveal,' and 'Anxiety Deconstructor' without any privacy, consent, or sensitivity warning. This creates a real risk that users will feed highly sensitive journals, therapy-like reflections, or third-party personal data into the skill without understanding the confidentiality and emotional-safety implications.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The skill advertises broad multi-source note analysis, including --source all, but provides no upfront privacy warning that it may scan highly sensitive personal notes across multiple repositories and services. Because the analysis perspectives include intimate profiling themes such as personality, anxiety, values, and relationships, omission of a privacy warning increases the chance users expose confidential data or trigger overbroad collection without informed consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill defaults to an output mode that appends generated analysis back into the original note, but the skill description and surrounding guidance do not clearly warn users that running it may modify their files. In a note-analysis skill operating over personal knowledge bases, silent write-back can unexpectedly alter user content, create integrity issues, and persist sensitive AI-generated inferences into files that may later sync or be shared.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.