ClawHub
Back to skill

Security audit

douyin-chat-insight

Security checks across malware telemetry and agentic risk

Overview

This skill is a local chat-export analysis tool with disclosed file access and report output, with limited optional environment checks but no evidence of exfiltration or hidden automation.

Install only if you are comfortable giving the skill local access to chat exports you explicitly choose. Treat generated reports as sensitive because they can include message excerpts and participant names; redact before sharing. Be aware setup/run may display whether related skills or cloud ASR API-key environment variables are present, but the inspected code does not send those values anywhere.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Lp3

Medium
Category
MCP Least Privilege
Confidence
81% confidence
Finding
The skill advertises operational capabilities such as environment access, file read/write, and shell execution, but does not declare permissions explicitly. Even though the documented behavior is privacy-conscious and includes safety boundaries, undeclared capabilities reduce transparency and can enable unintended access to local files, secrets in environment variables, or command execution if the implementation drifts from the documented workflow.

Context-Inappropriate Capability

Medium
Confidence
96% confidence
Finding
The function enumerates other installed skills under multiple agent-specific directories and reports whether specific companion skills exist, plus whether related API keys are present. For a tool whose stated purpose is local chat-export analysis, this local-environment fingerprinting is not necessary and creates avoidable privacy exposure that could be used to profile the user’s setup or drive conditional behavior based on detected tooling.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The metadata explicitly says the skill analyzes exported group/private chat data, which is inherently likely to contain sensitive personal information, identifiers, and private communications. Even though this file is only metadata, omitting any privacy warning or handling disclaimer can mislead users about the sensitivity of the data they are asked to provide, increasing the risk of accidental exposure or inappropriate processing.

Env Variable Harvesting

High
Category
Data Exfiltration
Content
for n in names:
        out[n] = any((r / n).exists() for r in roots)
    out["dashscope_key_present"] = bool(
        os.environ.get("DASHSCOPE_API_KEY") or os.environ.get("BAILIAN_API_KEY")
    )
    return out
Confidence
97% confidence
Finding
os.environ.get("DASHSCOPE_API_KEY

Env Variable Harvesting

High
Category
Data Exfiltration
Content
for n in names:
        out[n] = any((r / n).exists() for r in roots)
    out["dashscope_key_present"] = bool(
        os.environ.get("DASHSCOPE_API_KEY") or os.environ.get("BAILIAN_API_KEY")
    )
    return out
Confidence
97% confidence
Finding
os.environ.get("BAILIAN_API_KEY

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.