update-llms

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill analyzes repository content to generate summaries, which could allow malicious instructions in those files to influence the agent.
  • Ingestion points: Reads llms.txt, READMEs, and other documentation files within the repository.
  • Boundary markers: The skill lacks instructions to treat file content as untrusted or to use delimiters to prevent instruction following.
  • Capability inventory: The agent has the capability to read the filesystem and write the llms.txt file to the root.
  • Sanitization: No sanitization or filtering logic is provided for the content extracted from repository files.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 06:50 AM
Security Audit — agent-trust-hub — update-llms