Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion docs/.index
Original file line number Diff line number Diff line change
@@ -1,5 +1,5 @@
nav:
- Introduction: index.md
- Home: index.md
- OASF: oasf
- Agent Directory Service: https://dir.agntcy.org/latest/
- Secure Low-Latency Interactive Messaging (SLIM): https://slim.agntcy.org/latest/
Expand Down
Binary file added docs/assets/img/lf-horizontal-white.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file added docs/assets/img/lf-stacked-white.png
Loading
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
142 changes: 103 additions & 39 deletions docs/index.md
Original file line number Diff line number Diff line change
@@ -1,60 +1,124 @@
# AGNTCY Origins
---
hide:
- navigation
- toc
---

AGNTCY began with Outshift by Cisco's vision of an [Internet of Agents](https://outshift.cisco.com/the-internet-of-agents)—recognizing that AI agents were being built in isolated silos, unable to collaborate across organizational boundaries. Their foundational white paper identified the critical infrastructure gap preventing agents from working together at scale.
<div class="agntcy-landing">

What started as a concept became reality in March 2025 when Outshift launched AGNTCY on GitHub with complete code, specifications, and services, alongside Galileo and LangChain as core maintainers. They built the discovery, identity, messaging, and observability components that agents need to find each other, verify their capabilities, and collaborate securely.
<section class="agntcy-hero">
<div class="agntcy-hero__inner">
<h1 class="agntcy-hero__title">AGNTCY Documentation</h1>
<div class="agntcy-hero__partner">
<span class="agntcy-hero__partner-text">part of</span>
<a
href="https://www.linuxfoundation.org/press/linux-foundation-welcomes-the-agntcy-project-to-standardize-open-multi-agent-system-infrastructure-and-break-down-ai-agent-silos"
target="_blank"
rel="noopener noreferrer"
>
<picture>
<source
media="(max-width: 59.9375em)"
srcset="assets/img/lf-stacked-white.png"
/>
<img
src="assets/img/lf-horizontal-white.png"
alt="Linux Foundation"
class="agntcy-hero__partner-logo"
/>
</picture>
</a>
</div>
<p class="agntcy-hero__tagline">
Open infrastructure for the Internet of Agents.
</p>
<p class="agntcy-hero__lede">
AGNTCY is an open source collective building the software components and services
that let agents discover each other, verify capabilities, communicate securely,
and collaborate across frameworks and organizational boundaries.
</p>
</div>
</section>

By July 2025, over 75 companies had joined the effort, leading to AGNTCY's donation to the Linux Foundation with Cisco, Dell Technologies, Google Cloud, Oracle, and Red Hat as formative members. What began as one company's vision became the community-owned infrastructure for the Internet of Agents.
</div>

## Vision
<section class="agntcy-resources" markdown="1">

Agentic AI will accelerate all of human work. Enterprises need to create agentic workflows and applications combining internal and third-party agents to fully leverage the power of AI, accelerate their business, and achieve significant productivity gains.
We believe that an open, interoperable Internet of Agents is the key to enabling the best possible path forward to accelerate innovation and create the most value for all participants, from builders to operators, developers to consumers across all industries and businesses.
<h2 class="agntcy-section-title" id="explore-documentation">Explore the Documentation</h2>

## Mission
<div class="grid cards" markdown>

We are an open source collective committed to build the Internet of Agents to be accessible to all. Our mission is to build a diverse, collaborative space to innovate, develop, and maintain software components and services that solve key problems in the domain of agentic workflows and multi-agent applications.
- :material-folder-search-outline:{ .lg .middle } **Agent Directory Service**

## Capabilities
---

Based on advanced protocols, frameworks, and components, the goal of IoA software infrastructure is to enable and simplify the creation of multi-agent applications through the following steps:
Federated registry for publishing, verifying, and discovering agents and multi-agent applications.

1. **Discover**: Find and evaluate the best agents for the job.
1. **Compose**: Connect agents into effective workflows across any framework or vendor.
1. **Deploy**: Run multi-agent systems at scale, securely.
1. **Evaluate**: Monitor performance and improve efficiency and efficacy over time.
[:octicons-arrow-right-24: Directory docs](https://dir.agntcy.org/latest/){:target="_blank"}

## Technical Objectives
- :material-message-fast-outline:{ .lg .middle } **Secure Low-Latency Interactive Messaging (SLIM)**

1. **Interoperability**: Establish a common protocol that enables AI agents from different vendors and platforms to communicate and work together efficiently.
2. **Security**: Ensure secure interactions between agents through robust authentication, authorization, and encryption mechanisms.
3. **Scalability**: Design a scalable architecture that leverages the cloud-native stack optimally, supporting a growing number of agents and interactions without compromising performance.
4. **Standardization**: Develop standardized data models and schemas to ensure consistent data representation and validation across the ecosystem.
---

## Core Components
Secure network-level messaging for multi-agent applications with pub/sub, streaming, and MLS encryption.

The initial set of IoA components and architecture is outlined below. This is a starting point - as new members join and bring their contributions, the collective will continue to evolve and expand the IoA architecture, components, and interfaces.
[:octicons-arrow-right-24: SLIM docs](https://slim.agntcy.org/latest/){:target="_blank"}

![IoA Stack](assets/ioa_stack.png)
- :material-database-outline:{ .lg .middle } **Open Agent Schema Framework (OASF)**

1. **[Open Agent Schema Framework (OASF)](./oasf/open-agentic-schema-framework.md)**: An OCI based extensible data model allowing to describe agents' attributes and ensuring unique identification of agents. OASF supports the description of A2A agents, MCP servers and can be extended to support other popular formats, such as Copilot agent manifests and many more. Current OASF repo can be found [here](https://github.com/agntcy/oasf), OASF schema documentation can be found [here](https://schema.oasf.outshift.com).
1. **[Agent Directory](https://dir.agntcy.org/latest/)**: Allows announcing and discover agents or multi-agent applications which are described using OASF. Any organization can run its directory and keep it in sync with others, forming the Internet of Agents inventory. Agent Directory supports A2A agent cards and MCP server descriptions among other data models.
1. **[Secure Low-Latency Interactive Messaging (SLIM)](https://slim.agntcy.org/latest/)**:
* **SLIM** (Secure Low-latency Interactive Messaging): A protocol that defines the standards and guidelines for secure and efficient network-level communication between AI agents. SLIM ensures interoperability and seamless data exchange by specifying message formats, transport mechanisms, and interaction patterns.
* **SLIM Nodes and SDK**: Offers handy secure (MLS and quantum safe) network-level communication services to a group of agents (typically those of a given multi-agent application) through SDK/Libraries. It extends gRPC to support pub/sub interactions in addition to request/reply, streaming, fire & forget and more.
1. **[Identity](./identity/identity.md)**: A system that leverages decentralized technologies to manage and verify the identities of Agents or Tools issued by any organization, ensuring secure and trustworthy interactions.
1. **[Observability and Evaluation](./obs-and-eval/observe-and-eval.md)**: Telemetry collectors, tools and services to enable multi-agent application observability and evaluation.
1. **Security**: Tools and services to trust and protect multi-agent applications.
---

[CoffeeAGNTCY](./coffee-agntcy/get-started.md) is a reference implementation demonstrating the core components that can be used to build multi-agent applications.
Extensible data model for describing agent attributes, skills, and capabilities across A2A, MCP, and more.

The following diagram shows a simplified architecture of the core components described above.
[:octicons-arrow-right-24: OASF overview](oasf/open-agentic-schema-framework.md)

![IoA Arch](assets/ioa_arch.png)
- :material-shield-key-outline:{ .lg .middle } **Identity**

## Benefits
---

* **Enhanced Collaboration**: By enabling seamless communication and data exchange, IoA fosters collaboration between AI agents, leading to more sophisticated and integrated solutions.
* **Improved Efficiency**: Standardized protocols and frameworks reduce the complexity of integrating diverse AI agents, resulting in faster development and deployment of AI-driven applications.
* **Increased Security**: Robust security mechanisms ensure that interactions between agents are secure, protecting sensitive data, and preventing unauthorized access.
* **Future-Proof Architecture**: The scalable and flexible design of IoA ensures that the ecosystem can grow and adapt to future advancements in AI technology.
Decentralized identity for agents and tools—identifiers, verifiable credentials, and policy-based access.

[:octicons-arrow-right-24: Identity overview](identity/identity.md)

- :material-chart-line:{ .lg .middle } **Observability and Evaluation**

---

Telemetry collectors, tools, and services for multi-agent observability and performance evaluation.

[:octicons-arrow-right-24: Observability overview](obs-and-eval/observe-and-eval.md)

- :material-test-tube:{ .lg .middle } **CSIT**

---

Continuous system integration testing for AGNTCY components and multi-agent workflows.

[:octicons-arrow-right-24: CSIT overview](csit/csit.md)

- :material-coffee-outline:{ .lg .middle } **CoffeeAGNTCY**

---

Reference implementation demonstrating how to build multi-agent applications with AGNTCY components.

[:octicons-arrow-right-24: Get started](coffee-agntcy/get-started.md)

- :material-book-open-page-variant-outline:{ .lg .middle } **Glossary**

---

Definitions for terms used across AGNTCY documentation.

[:octicons-arrow-right-24: Glossary](glossary.md)

- :material-hand-heart-outline:{ .lg .middle } **How to Contribute**

---

Guidelines for contributing documentation, code, and community improvements.

[:octicons-arrow-right-24: Contributing](contributing.md)

</div>
</section>
8 changes: 8 additions & 0 deletions docs/stylesheets/custom.css
Original file line number Diff line number Diff line change
Expand Up @@ -105,6 +105,14 @@
color: var(--md-typeset-a-color);
}

.md-copyright--centered,
.md-copyright--centered .md-copyright__highlight {
width: 100%;
text-align: center;
font-size: 0.75rem;
color: var(--md-default-fg-color--light);
}

html [data-md-color-primary=black] .md-nav--primary .md-nav__title[for=__drawer] {
background-color: var(--md-default-bg-color);
}
Expand Down
Loading
Loading