Documentation
Build with Ajeris.
How the platform works, how to extend it with your own MCP tools, and the API surface Ajeris exposes to channels like Alexa+. Start with the quickstart or browse by category.
Getting started
Platform
Platform overview
How Ajeris is put together: your agent, the surfaces you reach it through, memory, and tools.
Read →Memory model
How Ajeris remembers. What stays between conversations, what fades, and how to teach the agent what matters.
Read →Scheduled tasks
How Ajeris runs work for you in the background - reminders, recurring jobs, and proactive checks.
Read →Skills
How Ajeris remembers HOW to do things, not just facts. Skills are procedural memory: proven recipes the agent reuses.
Read →Integrations
Connect your MCP server
How to plug a Model Context Protocol server into Ajeris - protocol, auth, and the connect flow.
Read →Claude Code Connect
Drive your laptop's Claude Code from any channel. Text your agent on WhatsApp, it routes coding tasks to the Claude Code session running on your Mac, reports progress, and commits the result, all without opening your laptop.
Read →Channels
Trust
Security model
Encryption, per-account isolation, OAuth token handling, and the engineering commitments behind Ajeris's trust posture.
Read →Access Controls Policy
How Ajeris controls access to production systems and consumer data. Per-account isolation in the database, MFA, authenticated webhooks.
Read →Vulnerability Management Policy
Continuous npm audit, Dependabot, CVSS-based SLAs, honest risk acceptance.
Read →Data Retention Policy
What Ajeris keeps, how long, and how it deletes. Where the full retention schedule lives and how to delete your data.
Read →