ClawHub
Back to skill

Security audit

Pet Companion Journal

Security checks across malware telemetry and agentic risk

Overview

This is a local pet journal skill that writes and reads user-directed pet records on disk, with no evidence of network exfiltration, hidden execution, or destructive behavior.

Install only if you are comfortable storing pet profiles, photos or photo paths, health notes, and reminders locally under the configured data directory. Use PET_COMPANION_HOME for isolation on shared machines, and review records before sharing reports or summaries.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • System Prompt LeakageDirect Leakage, Indirect Extraction, Tool-Based Exfiltration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill instructs the agent to use local Python scripts, shell commands, environment variables, and read/write operations under a user directory, but it declares no permissions or capability boundaries. That mismatch is a real security issue because it can cause the agent runtime or reviewers to underestimate what the skill can access or modify, especially since it handles health notes, photos, and reminder data on disk.

Vague Triggers

Medium
Confidence
90% confidence
Finding
These trigger phrases for adding records are broad enough to match common conversational utterances such as '记一下今天洗澡' or '记个照片' without strong pet-specific context. That increases the chance of unintended skill invocation, which could cause accidental creation of pet journal entries, misrouting user intent, or inappropriate capture of user-provided content into this skill's storage flow.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The query and reminder triggers include highly ambiguous phrases like '最近有没有异常', '今天要给它做什么', and '过去一个月都记了什么', which can overlap with ordinary conversation or other assistant domains. In a local-first journaling skill that handles health records and reminders, accidental invocation could expose or summarize sensitive pet-related history when the user intended a different task.

Env Variable Harvesting

High
Category
Data Exfiltration
Content
raise SystemExit("clawhub.json version must match skill.json")

    with tempfile.TemporaryDirectory() as tmp:
        env = os.environ.copy()
        env["PET_COMPANION_HOME"] = str(Path(tmp) / "pet-data")

        print("[verify] creating pet profile")
Confidence
60% confidence
Finding
os.environ.copy()

Direct Prompt Extraction

High
Category
System Prompt Leakage
Content
Use `scripts/export_report.py` to produce a compact summary for a time range.

## Output Rules

- Keep responses warm, clear, and organized.
- For write actions, confirm what was saved and under which pet.
Confidence
85% confidence
Finding
Output Rules

VirusTotal

57/57 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.