ling-kong-ran/pisper

Multi-agent app for parallel sessions, tools, memory, workflows, desktop, terminal, and mobile.

What it solves

Pisper is a multi-agent application designed to manage complex AI interactions by allowing users to branch conversations like code branches in Git. It solves the problem of linear conversation history by enabling parallel exploration of different ideas from any point in a chat, while maintaining separate contexts, models, and permissions for each session.

How it works

Built on the Pi Coding Agent runtime, Pisper organizes AI agents into a system of independent sessions. It uses a "branching" mechanism where a new session can be derived from any completed turn, inheriting the context without altering the original conversation. To optimize performance and cost, it employs a "hot/cold" tool layering system and prompt cache diagnostics to maximize provider cache hits. It also supports the Model Context Protocol (MCP) for extending capabilities via external services.

Who it’s for

Developers and power users who use coding agents as productivity tools and require a way to organize multiple parallel AI workflows, manage complex project memories, and access their agents across desktop, terminal (TUI), and mobile devices.

Highlights

  • Conversation Branching: Create new sessions from any previous turn to explore alternative paths without losing the original thread.
  • Cros-Platform Access: Native apps for Windows, macOS, Linux, Android, and iOS, including a terminal-based TUI and P2P connectivity between mobile and desktop.
  • Self-Expanding Capabilities: The agent can write, validate, and install its own local plugins to add new skills on the fly.
  • Privacy-First Design: Data is stored locally by default, with sensitive information automatically masked and memory requiring user confirmation before being used for recall.
  • Visual Workflows: Includes a visual editor to turn repetitive tasks into automated processes.

Related

  • Project
    rusty4444/hermes-android

    A native Android client for the Hermes Agent that enables mobile access to chat sessions, memory, and task automation via a secure local or private network connection.

  • Project
    pipecat-ai/pipecat

    Pipecat is an open‑source Python framework for building real‑time voice‑first and multimodal AI agents. It offers plug‑and‑play support for dozens of speech‑to‑text, LLM, and text‑to‑speech services, composable pipelines, and built‑in multi‑agent orchestration. The ecosystem includes client SDKs (JS, React, Swift, Kotlin, C++, ESP32), a CLI for scaffolding projects, a UI component library, a debugger (Whisker), and many example bots. Ideal for developers and researchers who want to prototype or deploy voice‑centric conversational systems quickly.

  • Project
    OpenWhispr/openwhispr

    A privacy-first, open-source voice-to-text application providing local AI dictation, meeting transcription with speaker diarization, and AI-powered notes.

  • Project
    luoguoxiong/aipack

    aipack is a TypeScript‑first, lightweight AI‑agent framework. It includes its own runtime, a Tapable‑based plugin system, and built‑in adapters for 13+ LLM providers. Features include streaming vs. batch execution, persistent session storage, permission‑based tool approval, hybrid BM25 + vector long‑term memory, five‑stage context compression, multi‑agent orchestration (graph, pipeline, router, debate, map‑reduce), two‑way MCP integration, full observability (metrics, dashboard), and an offline evaluation suite. The project ships a CLI assistant, a library for embedding in apps, and example apps demonstrating real‑world use cases.

  • Project
    horang-labs/tessera

    Tessera is a local desktop/web app that lets you run multiple Claude Code, Codex, or OpenCode coding agents in isolated worktrees, manage them with a Kanban board, and handle the full Git workflow without leaving the UI. It supports parallel sessions, a readable chat view for terminals, image handling, mobile remote access, and a CLI for lead‑agent orchestration. Install via native binaries, npm, or Docker; licensed AGPL‑3.0.