Agent-Reliability-and-Evaluation-Lab — tool profile and history

Production reliability, benchmarking, and evaluation layer for MCP agent runtimes — protocol negotiation, A2A interop, circuit breakers, connection pooling, adaptive streaming, and autoscaling, with measured throughput/latency and a 109-test suite.

Active

Canonical source ↗

Language
Python
License
Apache-2.0
Created
Last activity
Topics
  • a2a-protocol
  • agent-to-agent
  • ai-agents
  • async-python
  • autoscaling
  • backpressure
  • benchmarking
  • circuit-breaker
  • connection-pooling
  • fault-injection
  • llm-agents
  • mcp
  • model-context-protocol
  • observability
  • opentelemetry
  • python
  • reliability-engineering
  • resilience

Current measurement

3GitHub starsstable
measured growth

Momentum uses available readings from the last 7 days: measured with at least two comparable readings, estimated otherwise.

Reading history

90 days · 90 maximum readings
DateValueMetric
3GitHub stars
3GitHub stars
3GitHub stars
3GitHub stars
3GitHub stars
3GitHub stars
3GitHub stars
3GitHub stars
3GitHub stars
3GitHub stars
1GitHub stars

Classifications

Domains
Type
MCP servers
Use cases

These entries declare the same topics. No similarity is inferred: only the shared topics are stated.