codebase-onboarding

Pass

Audited by Gen Agent Trust Hub on Nov 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from a user-provided codebase to generate onboarding documentation, which could allow maliciously crafted code or manifest files to influence the agent's output.
  • Ingestion points: The scripts architecture_mapper.py, onboarding_generator.py, and setup_validator.py read project files such as package.json, requirements.txt, and source code. Additionally, references/fact-gathering-and-patterns.md instructs the agent to run discovery commands like cat, grep, and find on the target directory.
  • Boundary markers: The skill does not use specific delimiters or instructions to the agent to ignore potentially malicious directions embedded within the analyzed codebase files during the documentation generation phase.
  • Capability inventory: The skill possesses capabilities for local file reading and system tool discovery (e.g., via shutil.which). It does not perform network operations or write to sensitive file system locations.
  • Sanitization: The analysis scripts use structured parsing (JSON, regex) for metadata but do not explicitly sanitize or escape natural language content extracted from the codebase before interpolating it into the final markdown documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 8, 2026, 03:14 PM