Login

Integration

Long-term memory for AstrBot

AstrBot is the multi-platform chatbot framework that runs on QQ, WeChat, Discord, and Telegram all at once. Its built-in memory is per-conversation. The Engram plugin adds two slash commands (`/remember`, `/recall`) plus optional silent auto-archive, so every message a user sends becomes a memory in their personal bucket.

Get an API key (free) All integrations →

Install

Three steps: sign up for an Engram API key, paste a BYOK LLM-provider key on /models, then drop the snippet below into AstrBot.

Three steps to memory in your agent

  1. Sign up. Free, no card. You'll land on a Getting Started page that walks the next two steps.
  2. Add your LLM key. Engram is BYOK. Paste an OpenAI / Anthropic / Groq / Together / Fireworks key and we'll route every extraction and query call through your provider. You pay your provider directly. We never see your inference.
  3. Paste the snippet below into your agent and restart it. Use Authorization: Bearer <api-key>with the API key from your portal.

astrbot_plugin_engram: multi-platform chatbot framework

Adds /remember and /recall slash commands plus optional auto-archive of every non-command message. Per-user, per-chat, or fixed bucket strategy. Source: github.com/lumetra-io/engram-astrbot.

  1. Install from the AstrBot WebUI: Plugins → Install from URL → paste https://github.com/lumetra-io/engram-astrbot → Install → Enable.
  2. Open Plugins → Engram Memory → Config and paste your API key, or set ENGRAM_API_KEY in the AstrBot env and leave the form blank.
  3. Pick a bucket strategy: per_user (default; per-sender memory),per_chat (per-session), or fixed (one shared bucket). Setauto_archive: true if you want every non-command message persisted in the background.

What you can do once memory's wired in

  • Give every user across Discord, QQ, and WeChat the same persistent personality and preferences
  • Auto-archive every non-command message so the bot can answer 'what did we talk about last week?'
  • Run per-chat buckets in group rooms and per-user buckets in DMs, controlled by the bucket strategy config
  • Build a memory-first companion bot where every fact the user shares is queryable across platforms

FAQ

What's the difference between `per_user` and `per_chat` bucket strategies?

`per_user` keys by sender ID (memories follow the human across rooms); `per_chat` keys by session ID (memories scope to the conversation, useful for group rooms). `fixed` puts everything in one shared bucket.

Does `auto_archive` store slash commands too?

No, slash commands are skipped to avoid double-storing. Only normal messages over `auto_archive_min_length` chars get persisted. Default min length is 8.

Where does the plugin store the API key?

In AstrBot's plugin config (UI form or `data/config/astrbot_plugin_engram_config.json`). Falls back to the `ENGRAM_API_KEY` env var if the form is blank.

Related integrations

LibreChat Self-hosted ChatGPT alternative. OAuth per LibreChat user (no shared key in librechat.yaml), with a shared-bearer alternative for single-user deploys. Read more → Paperclip Auto-archives every issue, agent run, and approval into Engram with no agent code or LLM tool calls. Read more → Open WebUI Self-hosted local-AI UI, connected via Open WebUI's own mcpo SSE→OpenAPI bridge on a local port. Read more →

Ship durable memory in AstrBot today

Free tier: 10K memories and 50K retrievals per month. No credit card. Same Engram backend powers all 41 integrations, so memories you write from one client are immediately queryable from the rest.

Start free See pricing