agentfield and agentica

These are complements: AgentField provides the infrastructure for deploying agents as scalable services while Agentica provides the TypeScript framework for building agent function-calling logic that would run within such services.

agentfield
90
Verified
agentica
77
Verified
Maintenance 25/25
Adoption 20/25
Maturity 22/25
Community 23/25
Maintenance 20/25
Adoption 17/25
Maturity 25/25
Community 15/25
Stars: 881
Forks: 134
Downloads: 14,952
Commits (30d): 136
Language: Go
License: Apache-2.0
Stars: 1,002
Forks: 58
Downloads: 1,010
Commits (30d): 15
Language: TypeScript
License: MIT
No risk flags
No risk flags

About agentfield

Agent-Field/agentfield

Framework for AI Backend. Build and run AI agents like microservices - scalable, observable, and identity-aware from day one.

Provides a control plane that routes agent calls through REST APIs with built-in structured output validation (Pydantic/Zod schemas), human-in-the-loop pause/approval workflows, and cross-agent discovery. Supports Python, Go, and TypeScript SDKs; agents auto-register with cryptographic identity and produce tamper-proof audit trails. Features async execution with webhooks, canary deployments with traffic splitting, and integrated memory (KV + vector search) without external dependencies.

About agentica

wrtnlabs/agentica

TypeScript AI AI Function Calling Framework enhanced by compiler skills.

Supports three integration protocols—TypeScript classes, Swagger/OpenAPI documents, and MCP servers—allowing developers to compose agents by simply declaring existing functions without manual schema writing. Uses compiler-driven schema generation and automatic JSON Schema conversion across LLM vendors (OpenAI, Gemini, Claude, DeepSeek, Llama), while a selector agent filters candidate functions to optimize token consumption and reduce hallucination errors during function calling.

Scores updated daily from GitHub, PyPI, and npm data. How scores work