Geoskill: 城市绿地公平性

• Outcome: Pass
• Latest audit: Jul 30, 2026 · 3:49 PM UTC
• Version: 1.0.0

Overview

This skill appears to be a purpose-aligned local geospatial analysis tool, with ordinary output-file writes and some dependency hygiene issues to consider.

Install only in an isolated environment, pin or review the dependencies before production use, and provide an explicit output directory. Treat the documented auto-download/cache mode cautiously because the documentation and script do not match.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• MCP Least Privilege: Underdeclared Capability, Wildcard Permission, Missing Permission Declaration
• MCP Tool Poisoning: Hidden Instructions, Unicode Deception, Parameter Description Injection
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration

Findings (7)

Lp3

Medium

• Category: MCP Least Privilege
• Confidence: 87% confidence
• Finding: The skill documents file-writing behavior via multiple generated outputs, yet no declared permissions are present. In an agent setting, undeclared write capability undermines least-privilege controls and can cause users or orchestrators to invoke a skill that creates files, manifests, caches, or reports without explicit authorization.

Intent-Code Divergence

Medium

• Confidence: 89% confidence
• Finding: The documentation introduces automatic remote downloading from Microsoft Planetary Computer, which expands the skill from local geospatial analysis into network-enabled data acquisition. That mismatch can bypass user expectations, cause unintended external requests, and introduce supply-chain and data provenance risks because inputs are no longer solely user-provided local files.

Description-Behavior Mismatch

Medium

• Confidence: 85% confidence
• Finding: The manifest frames the skill as green-space equity analysis, but the interface adds AOI selection, date-range handling, and cache-directory management that materially broaden operational scope. Scope expansion is dangerous in agent ecosystems because policy decisions may be made from the high-level manifest, while the real behavior includes additional filesystem and potentially remote-data capabilities.

Unpinned Dependencies

Low

• Category: Supply Chain

Content

geoskill-data-fetcher>=0.2.0

# Third-party dependencies used by this skill
numpy
rasterio

• Confidence: 95% confidence
• Finding: numpy

Unpinned Dependencies

Low

• Category: Supply Chain

Content

# Third-party dependencies used by this skill
numpy
rasterio

• Confidence: 91% confidence
• Finding: rasterio

Unpinned Dependencies

Low

• Category: Supply Chain

Content

# Requirements for urban-green-equity
# Shared data-download library (Planetary Computer / NASA POWER / OSM)
geoskill-data-fetcher>=0.2.0

# Third-party dependencies used by this skill
numpy

• Confidence: 96% confidence
• Finding: geoskill-data-fetcher>=0.2.0

Known Vulnerable Dependency: numpy — 10 advisory(ies): CVE-2014-1859 (Numpy arbitrary file write via symlink attack); CVE-2021-41495 (NumPy NULL Pointer Dereference); CVE-2021-33430 (NumPy Buffer Overflow (Disputed)) +7 more

Critical

• Category: Supply Chain
• Confidence: 82% confidence
• Finding: numpy

VirusTotal

65/65 vendors flagged this skill as clean.

Static analysis

No suspicious patterns detected.