AIAny
AI Client2025
Icon for item

Macro

Unifies team email, chat, docs, tasks, CRM, calls and AI agents into a single workspace with bidirectional @-linking and a shared, exportable team memory.

Introduction

Most teams lose context because conversations, documents, email and CRM live in separate systems; that fragmentation is exactly what Macro tries to reverse by colocating everything and synthesizing a nightly team memory that agents can act on.

What Sets It Apart
  • Unified, bidirectional graph: every email, message, doc, task and CRM record is @-linked both ways, so you can trace decisions from conversation → task → code without switching tools.
  • Team-level memory for agents: a nightly synthesis produces a markdown-based, exportable memory that external agents (via MCP) or selected LLMs can query or act on, improving retrieval and agent reliability.
  • Agent-first surfaces and MCP: built-in agent workflows and a near-complete MCP/tool surface let agents draft/send email, create tasks, edit docs, and run actions programmatically without manual context plumbing.
  • Real-time CRDT docs and developer-friendly stack: live markdown editing backed by CRDTs (Durable Objects), with a Rust backend and SolidJS frontend for responsiveness and self-hosting under AGPLv3.
Who it's for and tradeoffs

Great fit if you manage a small-to-medium engineering or product team that needs to collapse tools (email, chat, docs, tasks, CRM) into one searchable context and wants agents to take safe, context-rich actions on behalf of the team. Self-hosting and tight integrations make it attractive for teams that care about data control and auditability.

Look elsewhere if you need a lightweight single-purpose tool (e.g., just email or just CRM), strict cloud SaaS guarantees from a vendor-managed product, or if you cannot accept AGPLv3 licensing for production deployments; Macro is opinionated about collocated workflows and expects teams to buy into that model.

Where it fits

Positioned as an "operating system" for small to mid-sized teams that treat AI agents as first-class collaborators. It competes with stacks assembled from Slack/Notion/Linear/HubSpot by offering one bidirectional database, agent integration via MCP, and exportable markdown memory—trading bespoke best-of-breed pieces for tight contextual coherence.

Information

  • Websitegithub.com
  • OrganizationsMacro (macro-inc)
  • Published date2025/11/08

More Items

Hugging Face

Provides 5.5K+ self-contained data-analysis RL tasks: each row bundles a real tabular dataset, a question, and a deterministically-gradable gold answer. Verified from jupyter-agent notebooks; splits for training, held-out testing, and quick eval; intended for prompting, fine-tuning, and agent RL.

Hugging Face
AI Model2026

A 9B agentic multimodal SFT checkpoint distilled from Qwen3.5-9B for coding, general agent tasks, visual coding and cybersecurity. Provided by Xiaomi MiMo as a research seed (77.4B-token SFT mix) to bootstrap agentic RL and tool-use experiments.

Hugging Face
AI Model2026

Preview agentic language model for research and engineering workflows that turns research questions into executable, verifiable workflows via tool use and long-context reasoning; built on a 744B-parameter MoE (GLM-5.2) with MIT-licensed BF16 and FP8 checkpoints.