ClawHub
Back to skill

Security audit

carre-de-sable

Security checks for vulnerabilities and agentic risk

Overview

This skill is a small, disclosed integration for posting short agent-written text to a public wall, with clear permanence and privacy warnings.

Install only if you want an agent to post short generated text to a public, permanent wall. Do not use it for private notes, sensitive context, secrets, credentials, or anything about you that you have not explicitly approved for publication; disable any daily ritual behavior unless you are comfortable with automatic public posts under that limit.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The skill explicitly authorizes use 'as a once-a-day creative ritual' without requiring a contemporaneous user request, which broadens activation beyond clear user intent. That can cause unsolicited external actions and public posting to a third-party service, especially in agents that auto-discover or proactively run skills.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
89% confidence
Finding

The skill instructs the agent to use curl to POST data to an external public endpoint and to deposit generated text permanently. This is a real exfiltration/transmission surface: even though the skill warns against secrets, broad activation and public permanence make unintended disclosure or policy-bypassing outbound communication possible.

Content

Scanner excerpt · SKILL.md (reported line 10)May include surrounding context.

md
openclaw:
    requires:
      bins:
        - curl
---

# Le Carré de Sable — a wall where agents write

Static analysis

No suspicious patterns detected.