kalibrShip agents that fix themselves. Kalibr learns what's working as your agents run in production and routes them around failures, degradations, and cost spikes...
Install via ClawdBot CLI:
clawdbot install devonakelley/kalibrKalibr hooks into your OpenClaw agent's runtime, reads live telemetry, and routes to the optimal execution path (model + tool + parameters) — preventing failures, degradations, and cost spikes before they impact users.
openclaw plugins install @kalibr/openclaw
openclaw config set plugins.entries.kalibr.config.apiKey "your-kalibr-api-key"
openclaw config set plugins.entries.kalibr.config.enableRouting true
openclaw gateway restart
Get your API key: https://dashboard.kalibr.systems
Telemetry mode (enableRouting: false) — instruments every LLM call with zero behavior changes. Full visibility into success rates, latency, cost, and failure patterns across providers.
Routing mode (enableRouting: true) — everything in telemetry mode, plus autonomous rerouting to the optimal model + tool + parameter combination. Uses Thompson Sampling + Wilson scoring. Detects provider degradation via 10% canary traffic and reroutes before users are affected.
Agents running in production with Kalibr see near 100% success rates even during provider incidents and degradations.
Generated Feb 25, 2026
A customer service AI that handles thousands of daily inquiries. Kalibr ensures the chatbot maintains near 100% uptime by automatically routing requests away from degraded LLM providers, preventing service interruptions during peak hours or provider outages.
An agent that processes quarterly financial reports and generates executive summaries. Kalibr optimizes for accuracy and cost by selecting the most reliable model for complex financial data while avoiding expensive providers for simpler tasks, ensuring consistent quality within budget.
A medical AI that helps prioritize patient symptoms based on severity. Kalibr's routing prevents failures by switching to backup models when primary providers experience latency spikes, ensuring critical health assessments are never delayed due to technical issues.
An automated moderation agent that screens user-generated content for policy violations. Kalibr uses canary traffic to detect when moderation models degrade in accuracy, automatically rerouting to maintain consistent detection rates without manual intervention.
An AI assistant that analyzes contracts and legal documents for risk assessment. Kalibr balances high-accuracy models for critical clauses with cost-effective options for routine sections, optimizing both precision and operational expenses.
Kalibr offers tiered monthly subscriptions based on request volume and features. Enterprise plans include advanced routing algorithms, dedicated support, and custom integration services, creating recurring revenue from businesses requiring reliable AI operations.
Customers pay per routed API call or per million tokens processed. This aligns costs with actual AI usage while providing savings through Kalibr's optimization, appealing to companies with variable workloads who want to control expenses.
Annual enterprise licenses include on-premise deployment, custom model integration, and service level agreements. This model targets large organizations with strict compliance requirements who need guaranteed reliability and dedicated configuration support.
💬 Integration Tip
Start with telemetry mode first to establish baseline performance metrics before enabling routing mode, and ensure your KALIBR_API_KEY is properly configured in environment variables.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Search and analyze your own session logs (older/parent conversations) using jq.
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linking related objects, enforcing constraints, planning multi-step actions as graph transformations, or when skills need to share state. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", entity CRUD, or cross-skill data access.
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vibe-coding ready.
Headless browser automation CLI optimized for AI agents with accessibility tree snapshots and ref-based element selection