[Labs] Budget enforcement and token metering skill for OpenClaw agents. Tracks AI spend via Revenium, enforces configurable budget guardrails, and reports usage automatically, so agents never silently blow through your token budget.
This repository also configures its own agents. See what openclaw-revenium tells them →
Skill Claude Code
Implementation blueprint from spike experiments for adding optional NemoClaw/OpenShell support to the Revenium OpenClaw skill. Requirements, proven patterns, and verified knowledge. Auto-load during NemoClaw/OpenShell implementation work.
Skill Claude CodeCodex
MANDATORY guardrail check BEFORE EVERY OPERATION — read guardrail-status.json first, always, no exceptions. Enforces Revenium guardrails-native budget rules, warns on threshold exceedance, halts autonomous agents on guardrail block, and meters usage into Revenium.