hackathon-test-generator
Pass
Audited by Gen Agent Trust Hub on Mar 12, 2026
Risk Level: SAFEPROMPT_INJECTIONREMOTE_CODE_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection from data ingested from upstream skills in the automated pipeline. \n
- Ingestion points: Data enters through the
mvp_features,mvp_demo_flow, anddone_criteriafields inSKILL.md. \n - Boundary markers: There are no delimiters or explicit instructions to treat the input data as untrusted content, allowing potential injection of instructions into the generation logic. \n
- Capability inventory: The skill generates functional source code stubs in the
test_scaffoldsoutput. \n - Sanitization: No input validation or output escaping is implemented for the provided strings. \n- [REMOTE_CODE_EXECUTION]: The skill performs dynamic code generation by creating test snippets (
test_scaffolds) based on the provided tech stack and feature descriptions. Because these snippets are generated from potentially untrusted external inputs, they could be exploited to deliver malicious code to the user's environment.
Audit Metadata