Skip to content

Latest commit

 

History

2 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 

Repository files navigation

Obyflow interactive terminal

Obyflow

AI-native, CLI-first observability for tracing, debugging, and understanding modern applications.

npm version PyPI version License: MIT


What we're building

Modern applications aren't just HTTP requests and database calls anymore — they're LLM calls, vector-store queries, and multi-step chains, all wired together across services. When something breaks, stitching together why usually means grepping logs across five dashboards by hand.

Obyflow captures structured events locally (SQLite — no external backend required), correlates them into full traces, and uses an LLM of your choice to turn raw evidence into a plain-English root-cause investigation.

npx obyflow start
npx obyflow investigate --since 1h
npx obyflow ask "why did checkout fail today?"

Quick start

npm install -g obyflow            # CLI
npm install @obyflow/node         # Node.js SDK
pip install obyflow-python        # Python SDK

Full docs, features, and repository layout live in the main repo README.

What's inside

Obyflow is a monorepo of focused packages rather than one monolithic SDK:

Package What it does
obyflow The CLI — init, start, traces, investigate, ask, incident, and more
@obyflow/core Shared event model, local storage, evidence graph, anomaly detection, correlation engine
@obyflow/node Node.js instrumentation SDK
obyflow-python Python instrumentation SDK (ASGI/WSGI middleware, LangChain callback)
@obyflow/llm-* Pluggable LLM providers — Anthropic, OpenAI, Gemini, Ollama
@obyflow/adapter-* LangChain and vector-DB adapters (Pinecone, Qdrant, Weaviate, Chroma, pgvector, Milvus)

All packages are versioned and released together for consistency across npm and PyPI.

Why Obyflow

  • Local-first — SQLite storage, zero infra to get started
  • Evidence-backed investigations — grounding validation flags LLM citations that don't match real evidence
  • "What changed" correlation — ties incidents to deployments, git commits, config and feature-flag changes
  • Built for AI-era stacks — first-class support for LLM calls, vector-store queries, and LangChain/LangGraph chains, alongside plain HTTP

Get involved


MIT Licensed · Built for developers debugging the next generation of applications

About

No description, website, or topics provided.

Resources

Stars

0 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors