piyaz
Active·★ 196·AGPL-3.0·Updated 2026-10-16
Piyaz is an agentic workspace where people and AI agents collaborate on projects, offering project management, agent coordination, and a shared context network. It solves the problem of agents lacking memory by creating tasks, planning implementations, and tracking progress, ensuring agents always have the right context for their work.
#AI Agents#Project Management#Context Management#Workflow Orchestration#Developer Tools#Full-Stack Development#PostgreSQL#Next.js
01
Features
01Living context network that maps project knowledge, decisions, and code relationships.
02Specialized context retrieval interface with four distinct shapes for different agent tasks (summary, working, planning, agent).
03Manages the full project lifecycle: Brainstorm > Decompose > Refine > Plan > Execute > Track.
04End-to-end task orchestration (Composer) that dispatches subagents for research, planning, implementation, and review.
05Interactive web UI with Structure and Graph modes for task management and visualizing the context network.
02
Compatibility
Claude Code
Agent Platform
Verified via docs
Codex
Agent Platform
Verified via docs
Cursor
Agent Platform
Verified via docs
Antigravity
Agent Platform
Verified via docs
Bun
Runtime
Verified via docs
PostgreSQL
Database
Verified via docs
03
Quick start
1
$ git clone [email protected]:FrkAk/piyaz.git
2
$ cd piyaz
3
$ bun install --production
4
$ cp .env.local.example .env.local
5
$ bun run db:setup
6
$ bun run build
7
$ bun run start
04
Use cases
↳Kickstarting new projects by brainstorming ideas and decomposing them into a structured task graph.
↳Onboarding existing codebases by reverse-engineering a task graph from code and Git history.
↳Seamlessly continuing ongoing projects with agents picking up tasks with appropriate context.
↳Driving specific tasks from research to pull request and merge using the Composer (Claude Code only).
↳Refining project scope and task details mid-flow or performing strategic reviews of the task graph.
05
Alternatives
agentic-cursorrules★ 646
A practical approach to managing multiple AI agents in Cursor through strict file-tree partitioning and domain boundaries.
pydantic-deepagents★ 1.1k
Python Deep Agent framework built on top of Pydantic-AI designed to help you quickly build production-grade autonomous agents with planning, filesystem operations, subagent delegation, and skills.
paneflow★ 87
Local-first Rust/GPUI workspace for running coding agents side by side: real terminal panes, live status, worktree review, read-only MCP, and local orchestration.
voidly-pay★ 10
Off-chain credit ledger + hire marketplace for AI agents. Ed25519-signed envelopes, atomic settlement, hire-and-release escrow. https://voidly.ai/pay
Comments
Log in to leave a comment
No comments yet. Be the first!