Documentation
Guides and tutorials to get the most out of gjalla: from CLI setup through architecture analysis and GitHub integration.
Quickstart
Core Concepts
- How gjalla works Start Here The whole model on one page: shared memory, rules, skills, system truth, and the loop that keeps them accurate.
- Rules and guidelines for your agents Plain-language rules, skills, and context that install into every agent's native machinery as real gate behavior.
- Shared agent memory Personal, team, system, and project memory: access, inheritance, lifecycle, and agent workflows.
- Attestations and telemetry What every agent reports back at a commit, how to record it, and what the accumulated record buys you.
Using gjalla
- Workspace tour Platform Tour of the portal: Overview, Activity, Specs & Docs, Steering, Automations, costs, and custom docs.
- gjalla CLI Guardrails Production-grade from every coding agent. Your standards, installed once, followed everywhere.
- Reviews & automations GitHub Integration Independent review on every pull request, plus configuring what runs and the reports you get.
- MCP server For AI Agents Connect the hosted MCP server by URL or run it locally, plus the session-start protocol and full tool reference.
Advanced
Can't find what you're looking for? Email support@gjalla.io.