deploy-staging

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (MEDIUM): The skill executes local shell scripts ./scripts/deploy-staging.sh and ./scripts/rollback-staging.sh. While consistent with the skill's purpose for deployment, this allows for the execution of any commands contained within those files with the agent's current privileges. This represents a risk if the repository content is not fully trusted.
  • [EXTERNAL_DOWNLOADS] (LOW): The skill uses curl to perform a health check on https://staging.example.com/health. This is a network operation to a non-whitelisted domain, though it is used here as a placeholder for a status check.
  • [PROMPT_INJECTION] (SAFE): No instructions designed to override the agent's system prompt or bypass safety guardrails were detected.
  • [DATA_EXFILTRATION] (SAFE): No attempts to access sensitive files (like SSH keys or credentials) or exfiltrate private data were found.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:06 PM