limits-energy
Pass
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill uses a bash script (
hooks/energy-gate.sh) to implement a user-defined energy budget. The script reads from local configuration files (~/.config/ai-limits/config) and logs (~/.config/ai-limits/state/energy-ledger.log). All logic is performed locally without network operations or external dependencies. - [SAFE]: Information regarding energy anchors (0.24 Wh and 0.34 Wh) is sourced from public vendor reports (Google and OpenAI) and is documented transparently in the skill instructions.
- [SAFE]: The budget gate mechanism includes an override feature (
state/override-energy-budget) that requires a deliberate user action, ensuring the agent does not autonomously block user intent without consent.
Audit Metadata