eliteforge-frontend-onboarding
Pass
Audited by Gen Agent Trust Hub on May 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes local shell scripts (
scripts/coverage_check.shandscripts/stitch_manifest.sh) to perform file system auditing, inventory creation, and ZIP extraction. These operations are executed within the project directory or temporary folders using standard utilities. - [PROMPT_INJECTION]: The skill processes untrusted HTML and image data from external prototype exports, which establishes an indirect prompt injection surface. Maliciously crafted prototype files could attempt to influence the agent's code implementation behavior.
- Ingestion points: Prototype HTML (
code.html) and associated image files. - Boundary markers: The instructions do not specify any delimiters or warnings to the agent regarding potential instructions embedded within the untrusted source data.
- Capability inventory: The skill allows the agent to initialize Git repositories, create and modify source files, and execute internal scripts.
- Sanitization: There is no requirement for the agent to sanitize or validate the content of the prototype files before using them as implementation guidance.
Audit Metadata