ClawHub
Back to skill

Security audit

TinkerClaw Agent Sensei

Security checks for vulnerabilities and agentic risk

Overview

This skill is a reference guide for operating AI agents and does not install, schedule, execute, or persist anything by itself.

Install this as reading material, not as an automation package. Users should treat its cron, memory, messaging, and config advice as operational guidance to apply deliberately, with human review before enabling scheduled jobs, changing config, sharing data, or writing durable memory.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • 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 (1)

Session Persistence

Medium
Category
Rogue Agent
Confidence
80% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · references/field-guide.md (reported line 293)May include surrounding context.

md
**Don't:**

- Use `nohup` manually when systemd is enabled
- Use `--force` (it kills existing processes including systemd's)
- Put `pkill` in `ExecStartPre` (can kill the process systemd just spawned — suicide loop)

Static analysis

No suspicious patterns detected.