skills/b-open-io/prompts/hammertime/Gen Agent Trust Hub

hammertime

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes local shell commands, specifically git status --porcelain, git log, and git ls-files, to inspect the repository state when the check_git_state feature is enabled for a rule. This is a documented functional requirement for determining if changes have been committed or pushed.
  • [EXTERNAL_DOWNLOADS]: The system leverages an external LLM (Anthropic's Haiku model) to perform secondary verification of ambiguous behavioral violations. This involves sending response text to a well-known AI service for scoring, which is a core component of the behavior enforcement feature.
  • [DATA_EXPOSURE]: The skill reads and writes configuration and state files within the user's home directory (specifically ~/.claude/hammertime or ~/.core/hammertime). It stores behavioral rules, session iteration counts, and debug logs locally to maintain persistence across agent sessions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 03:40 AM