project-init

Pass

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses standard command-line tools such as ls, mkdir, grep, and the GitHub CLI (gh) to perform its initialization tasks. These operations are limited to the local project directory and the user's GitHub account.
  • [EXTERNAL_DOWNLOADS]: No external downloads or remote scripts were detected. The skill interacts solely with the user's GitHub environment via the authenticated gh CLI.
  • [DATA_EXFILTRATION]: The skill reads project-specific files like okr.md and tasks.md to avoid overwriting existing data. It does not access sensitive system files (e.g., .ssh, .env, .aws/credentials) or send data to unauthorized external domains.
  • [PROMPT_INJECTION]: No malicious prompt injection patterns or attempts to bypass agent safety guidelines were found. The skill follows a structured, multi-phase guided workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 25, 2026, 04:36 PM
Security Audit — agent-trust-hub — project-init