Quickstart
1. Run the server
npx sideagent
It starts on http://localhost:8788 with an in-memory database (data is lost on exit) and prints the sign-in link to the log. For a real database set DATABASE_URL (Postgres with pgvector) and SIDEAGENT_SECRET (openssl rand -base64 32); see Server.
2. Create an agent
Open the dashboard, choose New agent, then follow its Setup page: connect and test the model, add knowledge, try it in the preview, list the origins that may embed the chat, and activate it. Copy its publishable key (pk_…). It is public.
3. Add the chat
import { AgentChat } from "sideagent";
<AgentChat publishableKey="pk_…" />
For signed-in users pass getToken; see Client.
Let your coding agent do it
Install the skill once, then prompt your agent: "Add an AI assistant to this app."
npx skills add egig/sideagent
The skill is also in the npm package at node_modules/sideagent/skills/sideagent/SKILL.md. If your agent has no skills support, ask it to read https://egig.github.io/sideagent/llms-full.txt and follow it. It is tested with eval cases in skills/sideagent/evals/ (pnpm eval:skill).
The agent answers from knowledge only. It cannot call your API or change data.
Other agent files
llms.txtandllms-full.txt: these docs for LLMs.AGENTS.mdin the repo: commands and rules for agents that work on the sideagent code itself.