business-logic-vulnerabilities
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and use various third-party security tools from GitHub repositories and external sites, including
AlphaSec/host_scan,tarunkant/Gopherus, and a sample zip bomb frombamsoftware.comfor testing purposes. - [COMMAND_EXECUTION]: The playbook contains numerous shell commands and scripts (using
curl,bash, andpython) designed to perform network operations, manipulate API requests, and conduct race condition testing against target systems. - [DYNAMIC_EXECUTION]: The skill includes Python code snippets for automated security testing tasks, such as performing OCR on captchas using
pytesseractand automated request hammering for identifying race conditions. - [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze potentially untrusted data from external target systems (e.g., API responses and HTML content) while maintaining capabilities like shell execution and network access.
- Ingestion points: Target website data and API responses during vulnerability assessment.
- Boundary markers: None specified in the instructions for handling untrusted target data.
- Capability inventory: Network requests via
curl, shell execution, and script creation/execution. - Sanitization: No instructions provided for escaping or validating external content before processing it within the agent's context.
Audit Metadata