astro-expert

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external code files and content collections for review, creating a vector for indirect prompt injection if those files contain adversarial instructions.
  • Ingestion points: The skill analyzes codebases and markdown files as specified in the target parameter and uses tools like Read and Glob to load content into the agent's context.
  • Boundary markers: There are no explicit instructions or delimiters used to separate the content being analyzed from the agent's operational instructions, increasing the risk of the agent obeying instructions embedded in the data.
  • Capability inventory: The skill is configured with a high-privilege toolset including Bash, Write, Edit, and Grep, which could be abused if an injection is successful.
  • Sanitization: No sanitization or filtering logic is present to identify or neutralize potential instruction injections within the analyzed content.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 03:57 PM