coral-review-source-spec

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from GitHub pull requests, source manifest files (YAML), and project documentation.
  • Ingestion points: Files located under sources/core/ or sources/community/, manifest.yaml, and README files in the target repository.
  • Boundary markers: The instructions lack specific boundary markers or delimiters to differentiate between external data and system instructions.
  • Capability inventory: The skill generates review findings based on analyzed code; however, the execution environment may include tools with file or network access that are not explicitly restricted.
  • Sanitization: There is no mention of sanitizing or escaping external inputs before the agent evaluates them.
  • [EXTERNAL_DOWNLOADS]: The skill references configuration and guidance from the author's official repository on GitHub (github.com/withcoral/coral).
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 10:08 AM
Security Audit — agent-trust-hub — coral-review-source-spec