poetry-rye-dependency-management

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of untrusted project data.
  • Ingestion points: The agent is directed to read pyproject.toml files using Glob, Grep, and Read tools, and to maintain state in .claude/context/memory/learnings.md.
  • Boundary markers: No explicit markers are provided to separate instructions from untrusted file content.
  • Capability inventory: The skill utilizes Bash, Write, and Edit tools which can execute commands and modify files.
  • Sanitization: There are no instructions for sanitizing or validating content from configuration files before tool execution.
  • [EXTERNAL_DOWNLOADS]: The skill documentation describes standard CI/CD workflows that download tools from well-known registries and use official GitHub Actions from trusted organizations like the actions organization. These references follow established best practices for development workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 08:55 AM