codebase-indexing

Pass

Audited by Gen Agent Trust Hub on Nov 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted codebase content to generate intelligence, creating a potential vector for malicious instructions embedded in project files.
  • Ingestion points: Reads and indexes project files and directory structures across the repository (SKILL.md).
  • Boundary markers: No specific delimiters or instructions are provided to the agent to treat indexed content as potentially untrusted data.
  • Capability inventory: Executes shell commands for tool setup and indexing, and interacts with the local file system (SKILL.md).
  • Sanitization: No content sanitization or source verification is performed on the indexed codebase content.
  • [EXTERNAL_DOWNLOADS]: Installs the third-party packages @colbymchenry/codegraph and repomix from the public npm registry.
  • [COMMAND_EXECUTION]: Utilizes shell commands to verify environment state, initialize graph databases, and update repository snapshots.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 14, 2026, 01:39 PM