Sidiora-Labs/hyperfide

The ai agent workflow tool that scales from small, simple projects to large, complex codebases.

What it solves

Hyperfide addresses the fragmentation between planning, coding, and documentation in large software projects. It acts as an agent-driven workflow engine that maintains a persistent, queryable state of a project—including code structure, task history, decisions, and documentation—so that AI agents can work on complex codebases with long-term memory and context awareness.

How it works

Hyperfide operates as a local CLI tool written in C11 that uses SQLite to store a comprehensive graph of your codebase. It indexes symbols, calls, and routes, and tracks "intent" through annotated comments called anchors. It manages a task-based workflow where agents claim tasks, receive relevant context (memories, symbols, and dependencies), and record outcomes. The system includes a "fleet" supervisor that can coordinate multiple agents across different branches or tasks, ensuring that work remains within defined scopes and is verified against the project's graph and history before merging.

Who it’s for

It is designed for developers and teams who use AI agents (like Claude Code or Codex) to build software and want to move beyond simple chat-based interactions toward a structured, verifiable, and persistent agent-driven development process.

Highlights

  • Persistent Memory: Stores decisions, constraints, and outcomes in a local database that survives agent session resets.
  • Context-Aware Retrieval: Provides agents with a ranked, budget-aware summary of relevant code, memories, and entry points in a single call.
  • Task-Driven Workflow: Uses a spec engine to define tasks with acceptance criteria, ensuring work is verified rather than just asserted as complete.
  • Local-First Architecture: Runs entirely on your machine with no telemetry, using a single binary and local SQLite storage for all indexing and state management.
  • Impact Analysis: Automatically calculates the "blast radius" of changes by tracing caller and callee edges across the codebase.

Related

  • Project
    Ingenimax/agent-sdk-go

    A Go SDK (and CLI) for building production‑ready AI agents with multi‑LLM support, persistent memory, tool integration, multi‑agent orchestration, guardrails, and enterprise features, configurable via environment variables or rich YAML files.

  • Project
    sleepinginsummer/agent-browser-cli

    A CLI tool that connects AI Agents to a real Chrome browser session, allowing them to reuse login states and perform precise page control and perception.

  • Project
    dream-num/dsh-univer-office

    A plugin for DeepSeek Harness that enables an AI agent to create, edit, and manage office documents like spreadsheets, presentations, and reports using natural language.

  • Project
    redis/mcp-redis

    Redis MCP Server is a Python‑based, Docker‑ready server that lets AI agents interact with Redis via natural‑language commands. It implements the Model Content Protocol, exposing tools for every Redis data type (strings, hashes, lists, sets, streams, JSON, vector indexes, pub/sub, etc.) and supports Azure Entra ID authentication. Install from PyPI or Docker, configure via CLI flags or env‑vars, and plug it into Claude Desktop, OpenAI Agents SDK, or any MCP client.