bazi-pillars-from-datetime

• Outcome: Pass
• Latest audit: May 29, 2026, 8:13 PM
• Version:
  1.0.1

Overview

This skill is a disclosed Bazi chart calculator with optional location lookup and caching that create privacy considerations but fit the stated purpose.

Install only if you are comfortable with birth-location names being resolved and possibly cached. To reduce exposure, provide longitude and latitude directly, set lookup_mode to local, avoid unnecessary geocoding API keys, and clear or relocate city_cache.json if cached locations are sensitive.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access

Findings (4)

Description-Behavior Mismatch

Medium

• Confidence: 89% confidence
• Finding: The skill performs online geocoding and writes a persistent cache even though its stated purpose is bazi derivation from datetime/timezone or local analysis. This scope expansion introduces undeclared outbound network access and local persistence of user location data, increasing privacy and data-handling risk beyond what a caller would reasonably expect.

Context-Inappropriate Capability

Medium

• Confidence: 84% confidence
• Finding: The code reads provider selection, cache paths, map paths, timeout, and API key material from environment variables for a skill that should mainly perform deterministic chart calculation. This adds hidden capabilities and deployment-time behavior changes, including use of external services and sensitive credentials, which makes the skill less transparent and harder to reason about safely.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: User-supplied location names are sent to third-party geocoding services without any disclosure or consent mechanism. Even if the data seems low sensitivity, location information can be personal data, and transmitting it externally creates privacy, compliance, and trust risks.

Missing User Warnings

Medium

• Confidence: 93% confidence
• Finding: The skill caches geocoded locations to disk without warning, creating undeclared local retention of potentially sensitive user location data. Persistent storage increases exposure to later unauthorized access, accidental leakage, or reuse beyond the original calculation request.

VirusTotal

63/63 vendors flagged this skill as clean.