🗂️ agentspace

Project: agentspace.directory is an open-source framework and discovery hub for LLM-agnostic, file-based AI agents. Built on the principle of Single-Source-of-Truth (SSoT), it allows developers to declare an agent's entire "soul"—its identity, logic, memory, and activity—within a single, portable directory of Markdown files.

By treating the file system as the primary orchestration layer, agentspace.directory eliminates fragmented configurations and provides a transparent, version-controlled environment where agents can be developed, tested, and shared regardless of the underlying language model.

For Developers & Power Users: These agents are for everyone—from complex coding tasks to family budgeting and personal memory curation.

Flexible Directory Pattern Single-Source-of-Truth

View on GitHub

Agent Space Hub

Dev & Test (The Builder)

Focus: Self-healing loops using apitests to fix code autonomously.

Procurement (The Buyer)

Focus: Business/Family governance; vetting vendors and simulating transactions via apistubs.

Second Brain (The Planner)

Focus: Proactive life management; scanning memories.md to trigger tracks for family events and travel (e.g., Malaga/Vilnius logistics).

🤖 Engine Agnostic Display

The entire agent behavior is declared in one single directory. It can be run seamlessly by different engines:

GPT Claude Gemini

Flexible Concept & Structure

The website facilitates a "Flexible Structure" approach to AI. Directory structure and agent logic depend on the set of projects and application, kind of projects, and main goals.

`instructions.md`

The "DNA" of the agent—system roles and logic.

`memories.md`

Shared long-term context and personal facts.

`tools.md`

Unified API and capability definitions.

`tracks/`

A subfolder for active state tracking (e.g., debug-xxx.md, feat-xxx.md).

Integrated Extensions

`apitests`

TDD for agents.

`apistubs`

Environment safety/mocking.

`apicron`

Proactive scheduling and time-based triggers.