Back to skill

Security audit

Nexus

Security checks across malware telemetry and agentic risk

Overview

This skill is a repository-mapping helper that discloses its local file reads and output files, with reasonable guardrails against secrets and unintended writes.

Install only from a source and revision you trust. Prefer the git clone route or verify the resolved npx package/version before running it. When using the skill, choose an authorized repository and private output directory, then review generated maps and context packs before sharing because file and symbol names can still reveal sensitive project information.

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
README.md:17
Finding
Unpinned Third-Party CLI Execution During Installation<![CDATA[ ## Vulnerability Details **File Location**: `README.md:17-23` **Vulnerability Type**: Supply-chain exposure through an unpinned npm CLI **Risk Level**: Medium ### Vulnerable Code ```markdown ## Install With the Agent Skills CLI: ```sh npx skills add AntreasAntoniou/nexus --skill nexus ``` ``` ### Technical Analysis The documented installation command invokes the third-party npm package resolved under the name `skills` through `npx` without specifying an exact version or integrity value. Depending on the local npm configuration and cache state, `npx` may download and immediately execute the package currently published under that name. Because the effective CLI implementation is not included in or pinned by this audited project, its behavior can change independently after the project has been reviewed. Compromise of the package publisher, publication of a malicious release, package-name confusion, or an unexpected upstream change could therefore introduce arbitrary code execution into the installation process. This finding concerns the documented installation method rather than malicious code in the bundled mapper. The alternative `git clone` command shown later in the README is more directly inspectable, although users should still verify the selected revision. ### Attack Path 1. An attacker compromises the npm account, publication process, or upstream package associated with the unversioned `skills` package, or otherwise causes a malicious release to be resolved. 2. A user follows the documented command: ```sh npx skills add AntreasAntoniou/nexus --skill nexus ``` 3. `npx` resolves and downloads the attacker-controlled or compromised package version. 4. The downloaded CLI runs with the permissions of the user performing the installation. 5. Malicious package code can access resources available to that user, such as local files, environment variables, developer credentials, agent configuration, and network services. Exploitation r ...[truncated 870 chars]
Remediation
<![CDATA[ ## Remediation Suggestions 1. Pin the installation CLI to an explicitly reviewed version: ```sh npx skills@<verified-version> add AntreasAntoniou/nexus --skill nexus ``` 2. Document the expected npm package publisher, canonical registry URL, and verified version so users can detect name confusion or publisher changes. 3. Where supported, require package integrity verification or distribute a lockfile recording the expected package version and integrity hash. 4. Prefer the auditable Git installation route and pin it to a release tag or commit: ```sh git clone --branch <verified-tag> --depth 1 https://github.com/AntreasAntoniou/nexus.git <destination> ``` 5. Advise users to inspect the package and resolved version before permitting `npx` to download and execute it. 6. Run installation with a non-privileged account and avoid exposing unrelated credentials or sensitive environment variables during installation. ]]>
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Lp3

Medium
Category
MCP Least Privilege
Confidence
89% confidence
Finding
The skill instructs the agent to read repository contents broadly and write multiple artifacts (`directory-map.md`, `context-pack.md`, and `sources.txt`) to disk, but it does not declare corresponding permissions. That mismatch is a real security issue because users and enforcement layers may assume the skill is read-only or lower privilege than it actually is, leading to unintended access to sensitive source paths or local file creation outside the repository.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
tests/test_mapper.py:12