ClawHub
Back to skill

Security audit

Write Coding Standards from File

Security checks for vulnerabilities and agentic risk

Overview

This skill is a documentation helper that may create or update project standards files and optionally fetch public style-guide references, with no evidence of hidden persistence, credential use, exfiltration, or destructive behavior.

Before installing, expect the skill to read only the files or narrow folder you provide, create a coding standards document by default, and possibly fetch public language style guides unless you set fetchStyleURL=false. Use outputSpecToPrompt=true for a read-only result, and review any proposed README, test-file, or source-file edits before allowing them.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill description promises not to modify analyzed source files unless edits are explicitly requested and confirmed, but the configuration defaults to creating a new standards file and can also insert content into README.md. Even if these are documentation files rather than analyzed source files, they are still repository modifications initiated without an explicit confirmation gate, creating a mismatch between declared behavior and actual write behavior.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill is supposed to derive standards from provided local files and folders, yet it enables external URL fetching by default. This expands the trust boundary, may leak contextual signals through network access, and allows externally sourced guidance to influence outputs without the user explicitly opting in.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

Generating and writing a test file goes beyond the advertised scope of producing a coding standards document and introduces additional repository modifications. This can surprise users, create unintended executable or CI-affecting artifacts, and violate least-privilege expectations for an analysis-oriented skill.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The instructions direct the agent to perform external fetches for many URLs without a clear user-facing warning or consent step. Unannounced network access is dangerous because it can disclose usage patterns or repository context indirectly and introduces dependency on untrusted external content during task execution.

Content

No source excerpt is available for this finding.