Skip to content
View parkavenue9639's full-sized avatar
6 followers · 8 following

Achievements

Achievement: Pull Sharkx2Achievement: YOLO

Achievements

Achievement: Pull Sharkx2Achievement: YOLO

Block or report parkavenue9639

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
parkavenue9639/README.md

A neofetch-style terminal card: an animated agent loop — observe, decide, guard, execute, ledger, repeat — with the llm called only when needed, and a prompt that types out agent CLI commands

Building observable, verifiable AI systems — from agent runtimes to scientific workflows.
专注 Agent 工程、开发者工具，以及可观测、可验证的 AI 工作流。

Selected Work · Open Source · Stack · Email

What I'm building

01 · ORCHESTRATE
Reliable agent systems with explicit lifecycle, routing, and human-control boundaries. 		02 · OBSERVE
Safe MCP tooling that turns runtime signals into compact, privacy-aware context. 		03 · VERIFY
Evidence-backed automation and reproducible AI workflows that fail clearly.

写得越少，bug 越少。自动化也应该可观察、可验证、可停止。

Selected work

jevloop

A Python agent runtime powered by TypeSafe Jev: guarded tool execution, isolated Docker sandboxes, and side-by-side LLM comparisons.

Python Agent Runtime Docker

OmniCell-Agent

An observable research agent for reproducible single-cell RNA-seq analysis.

LangGraph FastAPI React

loop-engineering

Maker-checker goal loops for bounded, evidence-backed coding-agent work.

Agent Skills Verification

external-agent-mcp

A local MCP runtime for asynchronous delegation to external CLI coding agents.

MCP Node.js CLI Agents

jaeger-mcp

A read-only, privacy-aware MCP server for querying and summarizing Jaeger traces.

Python MCP Observability

BaseGraph

A full-stack FastAPI + LangGraph starter with React, SSE, and persistent checkpoints.

FastAPI LangGraph SSE

More experiments → RAG, MLX, biomedical NLP, and developer-tooling explorations

Open source

stablyai/orca

An agent development environment for working with fleets of parallel coding agents.

TypeScript Electron MIT

agentscope-ai/PawBench

A benchmark for evaluating LLM × harness performance.

Python Agent Evaluation Apache-2.0

farion1231/cc-switch

A cross-platform desktop assistant for switching API providers across Claude Code, Codex, and other AI coding agents.

Rust Tauri MIT

Professional stack

Python, TypeScript, C, HTML, CSS, PyTorch, Django, FastAPI, React, PostgreSQL, Docker, Linux, macOS, Git, and Shell

  • AI & agents: LangGraph · LangSmith · PyTorch · NumPy · MLX · MCP
  • Backend & product: FastAPI · Django · React · PostgreSQL
  • Languages & web: Python · TypeScript · C · Shell · HTML5 · CSS3
  • Systems & workflow: Linux · macOS · Docker · Git · SONiC · Cursor

Activity

GitHub contribution graph

Let's connect

If you're working on agent reliability, MCP tooling, or applied AI, feel free to get in touch.

Pinned Loading

  1. ASKG ASKG Public

    Python 1

  2. MiniChatbot MiniChatbot Public

    Python 1

  3. BaseGraph BaseGraph Public

    Quick start by fastapi + langgraph

    Python 1

  4. external-agent-mcp external-agent-mcp Public

    JavaScript

  5. jaeger-mcp jaeger-mcp Public

    Python

  6. loop-engineering loop-engineering Public

    Loop-engineered goal mode for AI coding agents: visible previews, verification levels, checkpoints, delegation bounds, and terminal states.

    2