# wrtnlabs/agentica

TypeScript AI AI Function Calling Framework enhanced by compiler skills.

Repository: https://github.com/wrtnlabs/agentica
Canonical: https://ross.abutalabs.com/products/agentica
Homepage: https://wrtnlabs.io/agentica
Language: TypeScript
License: MIT
License Family: permissive
Topics: agent, ai, chatbot, llm-function-calling, multi-agent-system, openai, openapi, swagger, claude, llama, rag, retrieval-augmented-generation, agentic, agentic-ai, agentic-framework, typescript, function-calling
Last push: 2026-05-25T02:30:35+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 84, release rhythm 72, longevity 40
- inputs: {"age_days": 567, "days_push": 101, "days_rel": 105, "gap_med": 0.5, "n_releases_24m": 145}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1041, forks 62 (observed 2026-08-28T04:03:20.702554+00:00)

## What it is
Agentica is a TypeScript agentic AI framework specialized in LLM function calling, enhanced by compiler-driven validation via typia. Agents are built by simply listing functions from TypeScript classes, Swagger/OpenAPI documents, or MCP servers, without complex agent graphs or workflows.

## Use cases
- build an AI chatbot that calls my backend REST APIs from a swagger document
- create an agent from TypeScript class functions without writing agent graphs
- turn an OpenAPI spec into a conversational shopping assistant
- connect an MCP server as tools for an LLM agent
- add validation feedback so the LLM fills function arguments correctly
- build a function-calling agent with OpenAI or Claude in Node.js
- scaffold an agentic AI project with a CLI boilerplate

## When to choose
- you are a TypeScript developer wanting to build LLM agents from existing classes or OpenAPI/Swagger APIs
- you want function-calling agents without designing complex workflow graphs
- you need MCP server integration as agent tools
- you want compiler-validated function arguments with automatic retry feedback

## When to avoid
- you need a non-TypeScript/JavaScript stack
- you require fine-grained control over multi-agent orchestration graphs
- your project depends on Yarn Berry, which is unsupported
- you need a framework with a large mature ecosystem beyond early-stage tooling

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, rag, chatbot, sdk, mcp
- domain: large-language-models, chatbots, web-development, developer-tools
- platform: cross-platform
- tags: function-calling, openapi, swagger, typia, agentic-ai, multi-agent, openai, anthropic, nestia, react-native, ai-agents, retrieval-augmented-generation, nodejs, typescript

## Member repositories
- wrtnlabs/agentica (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:20.702554+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T07:02:59.881709+00:00, confidence not recorded.
  - readme: https://github.com/wrtnlabs/agentica (fetched 2026-08-28T04:03:20.702554+00:00, sha 7d82fbda477a)
  - homepage: https://wrtnlabs.io/agentica (fetched 2026-08-29T13:04:10.742852+00:00, sha 610b15d3e317)
  - site_page: https://wrtnlabs.io/agentica/docs (fetched 2026-08-29T13:04:10.745379+00:00, sha 090b9c1c065b)
  - site_page: https://wrtnlabs.io/agentica/docs/setup/cli (fetched 2026-08-29T13:04:10.747541+00:00, sha 9474ce364275)
  - site_page: https://wrtnlabs.io/agentica/docs/setup/manual (fetched 2026-08-29T13:04:10.749282+00:00, sha d92d684a5c46)
  - site_page: https://wrtnlabs.io/agentica/docs/setup/react-native (fetched 2026-08-29T13:04:10.751184+00:00, sha e4bcddead278)
  - site_page: https://wrtnlabs.io/agentica/docs/concepts/function-calling (fetched 2026-08-29T13:04:10.752990+00:00, sha 94d784a4247d)
  - site_page: https://wrtnlabs.io/agentica/docs/concepts/compiler-driven-development (fetched 2026-08-29T13:04:10.754940+00:00, sha 78497adbb305)
  - site_page: https://wrtnlabs.io/agentica/docs/concepts/document-driven-development (fetched 2026-08-29T13:04:10.757876+00:00, sha 79f51d372ac8)
  - site_page: https://wrtnlabs.io/agentica/docs/core (fetched 2026-08-29T13:04:10.760506+00:00, sha b284f2e81d87)
- Data as of 2026-08-30T08:39:29.467469+00:00.