Foglamp is the missing observability layer for AI agents, designed specifically for the Vercel AI SDK. With just two lines of code, it instruments every generateText and streamText call to provide deep visibility into cost, latency, token usage, distributed traces, evaluations, and alerts. Key features include:
- Cost Intelligence: Track spend per model, agent, or customer.
- Distributed Traces: Waterfall view of every run with exact prompts and responses.
- Evals: Score production traffic with code checks and LLM judges.
- Alerts: Threshold rules on cost, latency, and error rate.
- Per-Agent Monitoring: Spans, latency, and spend for each agent.
Foglamp helps developers catch regressions early, optimize costs, and ensure quality before users notice issues. It is open-source and available on GitHub.
