skills/mukul975/anthropic-cybersecurity-skills/detecting-insider-threat-behaviors/Gen Agent Trust Hub
detecting-insider-threat-behaviors
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is a well-structured collection of documentation, templates, and scripts for threat hunting. Its primary purpose is to help analysts detect insider threats through log analysis.
- [DATA_EXFILTRATION]: No evidence of unauthorized data transmission was found. The Python scripts (
scripts/agent.pyandscripts/process.py) process local log files and output analysis results to the local filesystem. There are no network requests or external data transfers. - [COMMAND_EXECUTION]: The scripts do not use dangerous functions like
os.system,subprocess.run,eval, orexecto run arbitrary shell commands. Logic is implemented using standard programmatic controls. - [REMOTE_CODE_EXECUTION]: There are no patterns involving the download and execution of remote scripts. The skill does not define external dependencies or install third-party packages.
- [PROMPT_INJECTION]: The skill instructions do not contain attempts to override agent safety guidelines or manipulate core behavior. The content is strictly professional and oriented toward cybersecurity analysis.
Audit Metadata