ClawHub
Back to skill

Security audit

amazon-daily-market-radar

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed Amazon market-monitoring tool that uses a ZooData API key, spends API credits, and stores local snapshots for daily comparisons.

Install only if you are comfortable sending tracked ASINs, competitor ASINs, keywords, category paths, and marketplace/date/filter values to ZooData on each run. Prefer setting ZOODATA_API_KEY in the environment or a secret manager, monitor credit use for scheduled runs, and delete the skill data folder if you want to reset or remove retained baselines.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (4)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The skill is presented as a narrowly scoped daily monitoring tool, but the embedded CLI surface exposes a much broader research and data-extraction toolkit, including raw review retrieval, keyword analytics, product research, and prompt-generation workflows. This scope expansion increases the risk of unauthorized or unexpected actions, higher-cost API consumption, and misuse in unattended automation because operators may grant trust and automation rights based on the narrower description.

Description-Behavior Mismatch

High
Confidence
91% confidence
Finding
The file is a shared full-surface CLI containing many workflows far beyond the advertised daily market radar purpose. Although there is an allowlist mechanism, the bundle still ships broad capability and relies on a separate manifest for confinement; if that manifest is absent in this skill context, the skill exposes unintended operations that can trigger additional data collection and external API calls outside user expectations.

Context-Inappropriate Capability

Medium
Confidence
80% confidence
Finding
The code includes review prompt-generation and keyword-intelligence features unrelated to a cron-style daily monitoring skill. In this context, extra capabilities increase the attack surface and enable collection or transformation of broader user/market data than the skill description suggests, which is risky for an unattended automation tool.

Session Persistence

Medium
Category
Rogue Agent
Content
print("    export ZOODATA_API_KEY='hms_live_yourkey'", file=sys.stderr)
    print("", file=sys.stderr)
    print("  Method 2: User-home config (persistent, shared across all skills)", file=sys.stderr)
    print("    mkdir -p ~/.zoodata && chmod 700 ~/.zoodata", file=sys.stderr)
    print('    (umask 077; echo \'{"api_key":"hms_live_yourkey"}\' > ~/.zoodata/config.json)', file=sys.stderr)
    print("    # keep the file private (0600) — it holds a bearer credential", file=sys.stderr)
    print("", file=sys.stderr)
Confidence
72% confidence
Finding
The tool supports persistent storage of a bearer API key in ~/.zoodata/config.json, and this skill is designed for unattended scheduled execution. In that context, persistent credentials increase blast radius if the host or account is later compromised, because the secret remains available across sessions and can be reused to access the external service.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.