lit-components

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEPROMPT_INJECTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill includes a review: never instruction in its frontmatter. This is a directive intended to override standard human-in-the-loop review processes, thereby reducing oversight of the agent's modifications to the codebase.\n- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to execute build and test scripts within a repository, creating a vulnerability surface where a compromised repository could execute malicious code during these phases.\n
  • Ingestion points: Source code and build configurations from the repository being modified (SKILL.md).\n
  • Boundary markers: Absent; the instructions do not specify any validation or isolation for the repository's scripts.\n
  • Capability inventory: Execution of arbitrary shell commands via compile, lint, and test scripts (SKILL.md).\n
  • Sanitization: Absent; the skill does not recommend checking the integrity of the environment or scripts before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 10:06 AM
Security Audit — agent-trust-hub — lit-components