Livestreaming to ATProtocol

Publish agent tool usage, responses, and reasoning to ATProtocol collections.

23|1|Updated Jan 24, 2026
One-click install
npx skills add https://github.com/just-cameron/central --skill livestreaming-to-atprotocol-just-cameron
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Livestreaming to ATProtocol
Source: https://github.com/just-cameron/central/tree/main/.skills/livestreaming-to-atproto
Command: npx skills add https://github.com/just-cameron/central --skill livestreaming-to-atprotocol-just-cameron

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Enable real-time publishing of an agent's tool usage, responses, and reasoning to ATProtocol collections.

Core Features & Use Cases

  • Real-time publishing: broadcast tool uses, responses, and reasoning to ATProtocol
  • Redaction and safety: automatically redact sensitive content before posting
  • Observability: support for glass-box cognition and external auditing of agent activity

Quick Start

Configure the hooks and run the livestream process to start streaming activity to ATProtocol.

Frequently Asked Questions about Livestreaming to ATProtocol

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I livestream agent activity to ATProtocol in real time?

To livestream agent activity to ATProtocol, configure Letta Code hooks to broadcast tool usage, responses, and reasoning to ATProtocol collections. This requires a running ATProtocol PDS endpoint, Letta API access, and valid environment credentials to begin streaming.

What is glass-box AI observability and how does it work with ATProtocol?

Glass-box AI observability provides external auditing of agent activity by transparently broadcasting tool calls, messages, and reasoning. Publishing this data to ATProtocol collections enables real-time transparency and external auditing of agent cognition and tool usage.

Can I broadcast Letta Code tool calls and reasoning to an ATProtocol PDS?

Yes, you can broadcast Letta Code tool calls, messages, and reasoning to an ATProtocol PDS. The Skill uses Letta Code hooks to capture agent activity and publish it to ATProtocol collections, provided you have a running PDS endpoint and API access.

Do I need a running ATProtocol PDS endpoint to publish agent activity?

Yes, a running ATProtocol PDS endpoint is required to publish agent activity. You also need Letta API access and configured environment credentials to authenticate and stream the agent's tool usage and reasoning data to the collections.

How is sensitive content handled when broadcasting agent responses to ATProtocol?

Sensitive content is automatically redacted before posting to ATProtocol. The Skill enforces content redaction and safety measures during the real-time publishing process to ensure potentially sensitive information is stripped from tool calls and reasoning before it reaches the collections.

What's the best way to set up real-time transparency for AI tool usage?

The best way to set up real-time transparency for AI tool usage is configuring Letta Code hooks to capture and broadcast agent activity. This approach publishes tool calls and reasoning to ATProtocol collections, establishing glass-box AI operations with built-in redaction for safe external auditing.