bug-bounty
Fail
Audited by Gen Agent Trust Hub on Jul 31, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides commands to install security tools from various external sources, including GitHub repositories (e.g.,
kiterunner,subzy) and package registries (pip3 install arjun,brew install trufflehog). - [COMMAND_EXECUTION]: The workflow relies on the execution of shell commands to perform reconnaissance and scan targets (e.g.,
subfinder,httpx,nuclei,ffuf). - [REMOTE_CODE_EXECUTION]: Although automated scanners flagged a network request to HackerOne as remote code execution, manual review confirms this is a connectivity check that discards the response body (
-o /dev/null). - [PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted content from targets (e.g., JS bundles, API responses) and tools without defining explicit boundaries or sanitization, while retaining capabilities for command execution and network access.
- [SAFE]: References to
book.hacktricks.xyzandhackerone.comare for legitimate security research and API interaction, respectively.
Recommendations
- HIGH: Downloads and executes remote code from: https://hackerone.com/graphql - DO NOT USE without thorough review
- Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata