skills/mukul975/anthropic-cybersecurity-skills/mapping-attack-paths-with-bloodhound-ce/Gen Agent Trust Hub
mapping-attack-paths-with-bloodhound-ce
Pass
Audited by Gen Agent Trust Hub on Aug 5, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads the BloodHound Community Edition
docker-compose.ymlfromghst.ly(a SpecterOps shortener) and references official SpecterOps and BloodHound repositories on GitHub. These are well-known and trusted sources for cybersecurity tooling. - [COMMAND_EXECUTION]: The skill provides various commands for running collectors (SharpHound.exe, AzureHound, bloodhound-ce-python) and deploying Docker containers. These commands are typical for the skill's intended purpose in a red-team or penetration-testing context.
- [REMOTE_CODE_EXECUTION]: No instances of remote code execution from untrusted sources were detected. The skill uses standard package managers and official download paths for its dependencies.
- [CREDENTIALS_UNSAFE]: The skill does not contain hardcoded credentials. It correctly instructs users to use environment variables (
BHE_SECRET) or command-line arguments for providing credentials to the collectors and the API helper script. - [PROMPT_INJECTION]: The skill body and metadata do not contain instructions that attempt to override agent behavior or bypass safety guidelines. The instructions are focused on providing a workflow for security analysis.
Audit Metadata