ClawHub
Back to skill

Security audit

Geoskill: Flood Impact Assessment

Security checks for vulnerabilities and agentic risk

Overview

This flood-impact skill is not malicious, but it overstates what it assesses and can generate reports using synthetic population data, so users should review it carefully before operational use.

Install only if you understand that auto-download mode synthesizes population data and the tool does not actually assess buildings or cropland despite advertising those layers. Do not use its reports for emergency or operational decisions without validating the data sources, pinning dependencies, and confirming the missing assessment layers are implemented.

SkillSpector

Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (8)

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding
The skill is presented as a flood impact assessment tool using real exposure layers, but the documented behavior includes auto-downloading remote data, synthesizing population data, and failing to actually assess some declared layers such as buildings and cropland. In a disaster-response context, this can mislead users into trusting incomplete or fabricated impact estimates, producing materially wrong decisions even without traditional code execution abuse.

Unpinned Dependencies

Low
Category
Supply Chain
Content
geoskill-data-fetcher>=0.2.0

# Third-party dependencies used by this skill
fiona
numpy
rasterio
shapely
Confidence
98% confidence
Finding
The dependency 'fiona' is unpinned, so installs may resolve to different versions over time, including newly introduced vulnerable or incompatible releases. In a geospatial skill that parses external spatial data, dependency drift increases supply-chain and reliability risk because security posture can change without review.

Unpinned Dependencies

Low
Category
Supply Chain
Content
# Third-party dependencies used by this skill
fiona
numpy
rasterio
shapely
Confidence
98% confidence
Finding
The dependency 'numpy' is unpinned, allowing uncontrolled version selection during installation. This can introduce vulnerable, breaking, or unexpected builds into an environment that may process untrusted raster or numeric inputs.

Unpinned Dependencies

Low
Category
Supply Chain
Content
# Third-party dependencies used by this skill
fiona
numpy
rasterio
shapely
Confidence
98% confidence
Finding
The dependency 'rasterio' is unpinned, so future installs may pull different versions with unknown security properties. Because this skill handles raster flood extent data, relying on floating versions increases exposure to parser-related vulnerabilities and supply-chain surprises.

Unpinned Dependencies

Low
Category
Supply Chain
Content
fiona
numpy
rasterio
shapely
Confidence
98% confidence
Finding
The dependency 'shapely' is unpinned, which permits non-reproducible installs and unreviewed upgrades. In geospatial processing, this raises risk from vulnerable parser/geometry-handling changes and makes incident response harder because deployed versions may differ.

Unpinned Dependencies

Low
Category
Supply Chain
Content
# Requirements for flood-impact-assessment
# Shared data-download library (Planetary Computer / NASA POWER / OSM)
geoskill-data-fetcher>=0.2.0

# Third-party dependencies used by this skill
fiona
Confidence
99% confidence
Finding
Using 'geoskill-data-fetcher>=0.2.0' sets only a lower bound, allowing any newer version to be installed without review. Since this shared library likely performs remote data access, an unexpected upgrade could introduce security regressions or behavior changes affecting the whole skill.

Known Vulnerable Dependency: fiona — 2 advisory(ies): GHSA-g4m4-9q4c-mfw6 (Fiona affected by CVE-2020-14152 related to madler-zlib); GHSA-q5fm-55c2-v6j9 (Fiona affected by CVE-2023-45853 related to MiniZip madler-zlib)

Critical
Category
Supply Chain
Confidence
91% confidence
Finding
The requirements file includes 'fiona' without constraining it to a patched release, and the scanner reports advisories affecting Fiona through bundled or linked compression components. This skill's context makes the issue more relevant because Fiona commonly processes externally sourced geospatial files, so vulnerable archive or data parsing paths could be reachable when handling untrusted datasets.

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
80% confidence
Finding
The file includes 'numpy' without any version constraint, and the scanner lists multiple historical advisories. This is a real dependency hygiene problem, but the actual exploitability is less certain here because several cited NumPy CVEs are old, disputed, or depend on specific local conditions; still, unpinned installation can select an affected build and numeric libraries are broadly exposed across the skill.

Static analysis

No suspicious patterns detected.