WeaveScope
Elixir native observability for AI agents
Last checked:
What is WeaveScope?
WeaveScope is a tracing and monitoring platform for Elixir agents that helps developers identify failing, slow, or expensive AI agent runs. It provides detailed visibility into model calls, tool calls, spans, latency, token usage, and costs through the BeamWeaver Elixir dependency, with minimal setup required.
WeaveScope pricing
Pricing model: Freemium
Free tier: up to 100K accepted events/month with no credit card required; paid tier pricing not specified in provided content
WeaveScope pros
- Free tier with no credit card required (up to 100K accepted events/month)
- One-dependency integration via BeamWeaver with minimal setup effort
- Comprehensive tracing showing model calls, tool calls, and spans in unified timeline
- Detailed monitoring of cost, token usage, latency (p95), error rates, and throughput
WeaveScope cons
- Elixir/Beam-only—cannot be used with other programming languages
- Early access/beta stage with incomplete feature set (alerting coming soon)
- Limited to Elixir agents—not suitable for other LLM application frameworks
Frequently asked questions about WeaveScope
How quickly can I integrate WeaveScope?
Very quickly—add the BeamWeaver dependency to mix.exs and configure one endpoint with your API key.
Do I need a credit card to start?
No, the free tier (up to 100K events/month) requires no credit card.
What can I trace with WeaveScope?
Model calls, tool calls, spans, latency, token counts, costs, error rates, and p95 latency across instrumented agent runs.
Is alerting available?
Not yet—alerting is coming soon and will be built on top of existing tracing and monitoring data.