eve-app-cli
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides documentation and code templates for creating CLI tools to simplify agent interactions with APIs.
- [DATA_EXFILTRATION]: The provided templates demonstrate how to access environment variables such as EVE_JOB_TOKEN and read local files for API requests, which are standard operations for a CLI tool within the Eve platform ecosystem.
- [COMMAND_EXECUTION]: The build script template includes standard file system operations to make the generated binary executable.
- [PROMPT_INJECTION]: The skill defines a pattern for agents to ingest CLI output, representing a surface for indirect prompt injection. Ingestion points: API responses and local JSON files. Boundary markers: The guide promotes structured JSON output via a --json flag to help delineate data. Capability inventory: The templates include network access and file read capabilities. Sanitization: JSON serialization is used for structured output.
Audit Metadata