ClawHub
Back to skill

Security audit

Clawd Docs V2

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward ClawdBot documentation helper that reads local documentation caches and fetches official docs when needed.

Install this if you want ClawdBot documentation lookup. Expect it to read ClawdBot docs files under ~/clawd/data and fetch pages from docs.clawd.bot when local snippets or cache are insufficient; the publisher should clean up the minor wording inconsistencies.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The skill repeatedly uses Slovak trigger/query examples such as "ako nastaviť telegram", "kto mi môže písať", and "Ako updatnúť ClawdBot?" while presenting them as the primary invocation/query patterns. There is no accompanying note that the skill is multilingual, optional for Slovak users, or intentionally region-specific, which can conflict with language/locale choice expectations.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

At L196, docs-cache is labeled 'Full page cache (future)', but earlier the skill instructs checking and reading from '~/clawd/data/docs-cache/' as a current operating step at L56-L68. This creates a direct contradiction about whether that cache exists and should be used.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
94% confidence
Finding

The v2.2.0 changelog says 'No external dependencies' at L221, yet the skill elsewhere documents 'Dependencies: web_fetch (Clawdbot core tool)' at L210 and repeatedly instructs use of web_fetch at L72-L80 and L162-L164. This is an active contradiction in the documentation about what the skill relies on.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.