repomix
Pass
Audited by Gen Agent Trust Hub on Oct 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of entire code repositories to create context for AI agents, which introduces a surface where malicious instructions embedded in the source code could potentially influence the agent's behavior.
- Ingestion points: The tool reads local file systems and remote GitHub repositories as specified in SKILL.md and references/usage-patterns.md.
- Boundary markers: Repomix provides structured output using XML tags or Markdown code blocks to separate individual files, helping the AI distinguish between different parts of the codebase.
- Capability inventory: The skill allows for repository packaging, file filtering, comment removal, and token counting, with the final output being provided to an LLM.
- Sanitization: The tool integrates Secretlint to detect and warn about sensitive information such as API keys and credentials before the output is shared.
Audit Metadata