# langwatch/better-agents

Standards for building agents, better

Repository: https://github.com/langwatch/better-agents
Canonical: https://ross.abutalabs.com/products/better-agents
Language: TypeScript
License: MIT
License Family: permissive
Topics: agent-framework, agent-testing, ai-agents, cli, dev-tool, lllm, llmops, mcp, python, typescript
Last push: 2026-06-03T17:12:53+00:00

## Health v2 (maintenance only)
Score: 70/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 85, release rhythm 71, longevity 35
- inputs: {"age_days": 501, "days_push": 91, "days_rel": 192, "gap_med": 1, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1553, forks 152 (observed 2026-08-28T04:05:02.856347+00:00)

## What it is
Better Agents is a CLI tool and a set of standards for scaffolding production-ready AI agent projects. It configures coding assistants like Claude Code or Cursor to be experts in a chosen agent framework, and generates a standardized project structure with scenario tests, versioned prompts, evaluation notebooks, and observability setup.

## Use cases
- scaffold a new AI agent project with best practices
- make my coding assistant an expert in LangGraph or Mastra
- add end-to-end scenario tests for agent behavior
- version and collaborate on LLM prompts
- set up evaluation notebooks for agent pipelines
- standardize agent project structure across a team
- bootstrap an agent project ready for production with observability

## When to choose
- starting a new agent project from scratch and wanting best practices baked in
- you use a supported framework like Agno, Mastra, or LangGraph
- your team needs standardized structure, prompt versioning, and agent testing
- you want your coding assistant guided by AGENTS.md and MCP configs

## When to avoid
- you have an existing agent codebase that doesn't fit the generated structure
- you need a runtime agent framework rather than a scaffolding tool
- you don't use AI coding assistants or the supported frameworks
- you need a framework-agnostic library to embed in your own code

## Facets
- artifact type: cli-tool
- maturity: active
- function: agent-framework, cli, testing, e2e-testing, developer-tools, mcp, prompt-engineering
- domain: large-language-models, developer-tools, testing
- platform: cli, cross-platform
- tags: agent-scaffolding, agents-md, scenario-testing, prompt-versioning, llmops, project-generator, coding-assistant, scaffolding, ai-agents, command-line, nodejs

## Member repositories
- langwatch/better-agents (main) score 70

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:02.856347+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 2026-08-30T04:30:09.453493+00:00, confidence not recorded.
  - readme: https://github.com/langwatch/better-agents (fetched 2026-08-28T04:05:02.856347+00:00, sha c761c4127e8c)
- Data as of 2026-08-30T08:39:29.467469+00:00.