skill-file-security
Warn
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs users to run
npx @netxeo/security-skill, which downloads and executes code from a remote NPM package. This package is not from a recognized trusted organization and performs automated modifications to the project environment.- [EXTERNAL_DOWNLOADS]: The installation process fetches the@netxeo/security-skillpackage from the public NPM registry, introducing an external dependency into the development workflow.- [COMMAND_EXECUTION]: The skill defines several slash commands (e.g.,/security-scan,/security-audit,/security-fix) that authorize the AI agent to execute shell commands and perform file system operations to analyze and modify the project's codebase.- [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves the AI agent scanning and processing project files (source code, configuration, and environment files). This creates an attack surface where malicious instructions embedded in those files could influence the agent's behavior. - Ingestion points: Project source files,
package.json,Dockerfile, and.envfiles are read by the agent during audits. - Boundary markers: The instructions do not specify any delimiters or safety warnings to ignore embedded instructions within the analyzed data.
- Capability inventory: The agent is granted capabilities to read files, write files, and execute shell commands.
- Sanitization: There is no mention of sanitization or validation of project content before it is passed to the LLM for analysis.
Audit Metadata