codeguard-crystal
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to read and process local Crystal source files and configuration files (shard.yml), creating a vulnerability surface where instructions embedded in project data could potentially influence agent behavior. \n
- Ingestion points: The agent reads the content of
.crsource files andshard.ymlproject files during the linting and formatting process. \n - Boundary markers: The instructions do not define specific delimiters or security headers to differentiate project content from agent instructions during ingestion. \n
- Capability inventory: The skill possesses the capability to execute shell commands (
ameba,crystal tool format) and write modifications to the local filesystem. \n - Sanitization: There are no documented mechanisms for sanitizing or validating the contents of project files before they are interpreted by the agent.
Audit Metadata