Back to skill

Security audit

Local Coding

Security checks for vulnerabilities and agentic risk

Overview

This skill is a coherent local coding assistant, with ordinary local-service and package-install risks users should understand before installing.

Install this only if you trust the ollama-herd package and the devices on your local fleet. Prefer a virtual environment, avoid running pip as root or administrator, consider pinning a reviewed package version, and remember that prompts, code snippets, traces, audio files, and image prompts may be handled by your local router and fleet logs.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:34
Finding

Unpinned Third-Party Package Installation

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, line 34
Vulnerability Type: Unpinned and unverified third-party dependency installation
Risk Level: Medium

Vulnerable Code

bash
pip install ollama-herd    # PyPI: https://pypi.org/project/ollama-herd/

Technical Analysis

The Quick Start instructions install ollama-herd directly from PyPI without pinning a reviewed version or verifying package hashes. The command can therefore resolve to whichever package release and transitive dependencies are current when the user runs it.

Python package installation may execute package build or installation logic with the privileges of the invoking user. Consequently, compromise of the package, its maintainer account, its publishing pipeline, or one of its dependencies could turn this documented installation step into arbitrary local code execution.

The package name and source are consistent with the Skill's declared functionality, and the audit found no evidence that the current package is malicious. The risk arises from mutable, unverified supply-chain inputs rather than demonstrated malicious behavior in this repository.

Attack Path

  1. An attacker compromises the ollama-herd PyPI project, a maintainer account, its release pipeline, or a dependency selected during resolution.
  2. The attacker publishes a malicious package release or dependency version.
  3. A user follows the Skill's Quick Start instructions and runs pip install ollama-herd.
  4. pip retrieves the attacker-controlled release because no version or cryptographic hash is constrained.
  5. Malicious build, installation, or imported runtime code executes on the user's system.
  6. The payload can access resources available to the account that performed the installation.

Impact Assessment

Successful exploitation could provide arbitrary code execution with the privileges of the user running pip. Potential scope includes reading o ...[truncated 812 chars]

Remediation
View remediation

Remediation Suggestions

  1. Pin the dependency to a specifically reviewed release, for example:
    bash
    python3 -m pip install "ollama-herd==REVIEWED_VERSION"
    
  2. Publish and verify cryptographic hashes using a locked requirements file and --require-hashes:
    bash
    python3 -m pip install --require-hashes -r requirements.lock
    
  3. Lock and review all transitive dependencies rather than constraining only the top-level package.
  4. Recommend installation inside a dedicated virtual environment or another isolated environment instead of the user's system Python environment.
  5. Document the package's authoritative provenance, release-signing or attestation process, and procedure for verifying artifacts.
  6. Advise users not to run the installation as root or with administrator privileges.
  7. Periodically review the pinned release and dependency set before publishing updated installation instructions.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 100)May include surrounding context.

bash
# Models loaded in memory
curl -s http://localhost:11435/api/ps | python3 -m json.tool

# All available models
curl -s http://localhost:11435/api/tags | python3 -m json.tool

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 103)May include surrounding context.

md
curl -s http://localhost:11435/api/ps | python3 -m json.tool

# All available models
curl -s http://localhost:11435/api/tags | python3 -m json.tool

# Recent coding request traces
curl -s "http://localhost:11435/dashboard/api/traces?limit=5" | python3 -m json.tool

External Script Fetching

High
Category
Supply Chain
Confidence
90% confidence
Finding

Remote code is downloaded and executed. This bypasses code review and could introduce malicious code.

Content

Scanner excerpt · SKILL.md (reported line 106)May include surrounding context.

curl -s http://localhost:11435/api/tags | python3 -m json.tool

Recent coding request traces

curl -s "http://localhost:11435/dashboard/api/traces?limit=5" | python3 -m json.tool

text

## Also available on this fleet

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 55)May include surrounding context.

Code review

bash
curl http://localhost:11435/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-coder-v2:16b",

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill is advertised as a local coding assistant, but it also documents general-purpose chat, image generation, and speech-to-text through the same service. This expands the accessible capability surface beyond the declared purpose, which can cause users or downstream agents to invoke unintended high-privilege or out-of-scope features.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 116)May include surrounding context.

Image generation

bash
curl http://localhost:11435/api/generate-image \
  -d '{"model": "z-image-turbo", "prompt": "developer workspace illustration", "width": 512, "height": 512}'

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The early documentation emphasizes code-focused, local-only behavior, but later sections reveal broader multimodal functionality on the same endpoint family. This inconsistency can mislead users and security tooling about what data types and operations the service actually supports, weakening informed consent and policy enforcement.

Content

No source excerpt is available for this finding.