ChatCLI
A unified AI command-line and server platform that connects 14 LLM providers with failover, runs coordinated autonomous agents behind a ReAct engine, and also serves as a gRPC server and Kubernetes operator.
ChatCLI connects fourteen LLM providers, including OpenAI, Anthropic, Bedrock, Google, xAI and Ollama, with intelligent error classification, exponential backoff and per-provider cooldown on failure. Fourteen built-in autonomous workers, twelve orchestration specialists plus a refiner and a verifier, are coordinated by a ReAct engine and run a seven-pattern quality pipeline combining Self-Refine, Chain-of-Verification, Reflexion, retrieval-augmented generation with HyDE, and Plan-and-Solve, composed through a thread-safe state machine with circuit breakers and hot reload.
A durable scheduler runs cron, wait-until and DAG jobs that survive process crashes, backed by a write-ahead log, snapshots and Prometheus metrics. Production features include gRPC with TLS 1.3, JWT and role-based access control, AES-256-GCM encryption, rate limiting and audit logging, and a Kubernetes operator adds seventeen custom resource definitions with an autonomous remediation pipeline. It extends through signed plugins, multi-registry skills and slash-command templates interoperable with several other coding-agent CLIs, and is written in Go under the Apache-2.0 license.