ClawHub
Back to skill

Security audit

home-assistant-causal-incident-analysis

Security checks for vulnerabilities and agentic risk

Overview

This skill is a narrow Home Assistant troubleshooting guide and does not include hidden code, persistence, or data-export behavior.

Before installing, be comfortable letting the agent inspect Home Assistant logs, entity history, automations, and registry details for a specific incident. Keep reports scoped and redact tokens, MAC addresses, registry identifiers, and unrelated household configuration as the skill itself instructs.

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 (1)

Context Leakage

High
Category
Data Exfiltration
Confidence
75% confidence
Finding

Code or instructions that leak agent conversation context to external services, potentially exposing sensitive user interactions.

Content

Scanner excerpt · SKILL.md (reported line 21)May include surrounding context.

md
- Add upstream trigger sensors and downstream telemetry only after finding a candidate path.
   - Request minimal responses without attributes when attributes are not needed.
   - Narrow the time window or split entities if output truncates.
6. Query the logbook for each state-changing control entity. Capture the context entity and context name; use them to attribute a transition to an automation when present.
7. Build a timestamped chain: upstream sensor change → automation decision → control switch transition → downstream device telemetry. Preserve gaps and polling delays instead of forcing exact simultaneity.
8. Separate evidence levels:
   - Direct: logbook attributes the transition to an automation.