ClawHub
Back to skill

Security audit

Grill With Docs

Security checks for vulnerabilities and agentic risk

Overview

This skill is a documentation assistant that reads project context and writes glossary/ADR documentation, with no hidden code execution or unrelated data access found.

Install this only if you want the agent to read your project context and create or update repository documentation such as CONTEXT.md and docs/adr files. Review generated docs before committing them, especially because updates may happen during the session.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (3)

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explicitly instructs the agent to create CONTEXT.md when none exists, which is a repository-modifying action performed during normal use without an explicit user-facing warning or consent checkpoint. In an agent setting, silent file creation can lead to unintended workspace changes, documentation drift, or modification of sensitive repositories when the user only expected conversational guidance.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The instruction to update CONTEXT.md inline during the session means the agent may repeatedly edit files as the conversation progresses, again without an explicit warning that the repository is being modified in real time. This increases risk because writes are incremental and easy for a user to miss, making accidental or undesired changes more likely than a single end-of-session review.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill promises to generate ADR files at the end of the session, which can create multiple new files in the repository without clearly warning the user ahead of time. Even though the content is documentation-oriented, unannounced file creation is risky in agent workflows because it can pollute repos, trigger automation, or commit unwanted decisions as if they were finalized artifacts.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.