hunt-spa-api
Fail
Audited by Gen Agent Trust Hub on Nov 12, 2026
Risk Level: HIGHCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: Employs shell commands such as curl for network requests and API testing, grep for identifying API routes and secrets within targets, and mkdir for local file management as seen in the multi-step play in SKILL.md.
- [EXTERNAL_DOWNLOADS]: Downloads target HTML resources and JavaScript application bundles from external, user-specified URLs (e.g., console.target.com) for local inspection.
- [INDIRECT_PROMPT_INJECTION]: Analyzes untrusted external web content which presents a vulnerability surface where malicious strings in target code could influence agent behavior. 1. Ingestion points: Remote HTML and JavaScript assets downloaded in SKILL.md Steps 1 and 2. 2. Boundary markers: No delimiters or ignore instructions used for ingested content. 3. Capability inventory: Shell command execution (bash) and network operations (curl). 4. Sanitization: Lacks sanitization or validation of the downloaded content before processing via grep.
Recommendations
- HIGH: Downloads and executes remote code from: https://console.target.com/ - DO NOT USE without thorough review
Audit Metadata