sessions

Manage persistent AI agent sessions and inspect SQLite traces in Ravi.

6|2|Updated Jan 29, 2026
One-click install
npx skills add https://github.com/filipexyz/ravi --skill sessions-filipexyz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: sessions
Source: https://github.com/filipexyz/ravi/tree/main/src/plugins/internal/ravi-system/skills/sessions
Command: npx skills add https://github.com/filipexyz/ravi --skill sessions-filipexyz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill solves the challenge of maintaining context and continuity for AI agents across multiple days, channels, and provider restarts by providing a robust, durable session management layer.

Core Features & Use Cases

  • Session Lifecycle Management: Create, rename, reset, and prune sessions to keep your agent environment clean and organized.
  • Multi-Channel Routing: Attach multiple chat sources to a single session to unify history and control output targets.
  • Operational Observability: Inspect SQLite traces to debug runtime incidents, provider handoffs, and delivery failures.

Quick Start

Use the sessions skill to list all active sessions and then inspect the trace of the main session to debug recent activity.

Frequently Asked Questions about sessions

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

FAQPage Schema
How do I manage persistent conversation state for AI agents across provider restarts?

You manage persistent conversation state by utilizing a durable session management layer that provides continuity for AI agents across multiple days, channels, and provider restarts. This prevents context loss during operational interruptions.

Can I route multiple chat channels to a single AI agent session?

Yes, you can route multiple chat channels to a single session using multi-channel subscription features. This unifies conversation history from multiple chat sources into one session and controls output targets.

How do I debug AI agent runtime incidents and provider handoffs?

You debug AI agent runtime incidents and provider handoffs by inspecting SQLite traces. Querying the local SQLite database allows you to review delivery failures and analyze provider interactions for troubleshooting.

Do I need a local Ravi daemon to manage AI agent sessions?

Yes, you need a local Ravi daemon to manage AI agent sessions. Performing stateful operations, session routing, and event analysis requires access to the local Ravi runtime environment and its associated SQLite database.

What is the best way to prune and reset AI agent sessions?

The best way to prune and reset AI agent sessions is through dedicated session lifecycle management operations. You can create, rename, reset, and prune sessions to keep your runtime environment clean and organized.