@tank/hook-creator

Pass

Audited by Gen Agent Trust Hub on May 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill documentation provides technical templates for executing shell commands using the ctx.$ API within agent lifecycle hooks.
  • Evidence: Examples in references/worked-examples.md demonstrate the use of shell execution for retrieving git differences and running code formatters.
  • [DATA_EXFILTRATION]: The skill requests broad read access to the local filesystem (**/*) in its configuration.
  • Evidence: The tank.json manifest declares filesystem: { "read": ["**/*"] }, which is intended to allow the agent to assist in analyzing project structure for hook creation.
  • [EXTERNAL_DOWNLOADS]: The documentation references various external development tools and formatters.
  • Evidence: references/worked-examples.md suggests the use of standard packages and tools such as prettier, black, and gofmt for automated formatting tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
May 31, 2026, 11:04 PM
Security Audit — agent-trust-hub — @tank/hook-creator