dep-egress
Pass
Audited by Gen Agent Trust Hub on Aug 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted third-party source code and package metadata, which presents a surface for indirect prompt injection. This attack vector is inherent to the skill's purpose as a code auditing tool.
- Ingestion points: Third-party source code files, package manifests (package.json, setup.py, pyproject.toml), and build/install scripts (build.rs).
- Boundary markers: The instructions require the agent to provide specific citations (file paths and line numbers) and to quote findings directly, which helps maintain structural context separation.
- Capability inventory: The skill utilizes standard file system reading and pattern matching capabilities.
- Sanitization: The workflow does not specify explicit sanitization or filtering of the external content being analyzed.
- [SAFE]: The skill includes a reference to a companion resource, SKILLmama, at a public GitHub repository (https://github.com/Magithar/SKILLmama). This is a legitimate resource from the skill author ('magithar') and does not involve unsolicited remote code execution or suspicious downloads.
Audit Metadata