feat: add qwe-qwe template — self-hosted AI agent#1390
Conversation
✅ Deploy Preview for easypanel-templates ready!
To edit notification comments on pull requests, go to your Netlify project configuration. |
Builds and pushes ghcr.io/deepfounder-ai/qwe-qwe:latest on every main push, and versioned tags (e.g. :0.18.4) on version tags. Required for Easypanel template (PR: easypanel-io/templates#1390). Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
|
Hi @dray-supadev, There are a lot of changes required on this PR.
Please make sure to follow the what's being followed by taking any of my PRs as a reference that are still open. |
…back PR review feedback (easypanel-io#1390): root of template should contain only meta.yaml and index.ts; logos and screenshots belong in an `assets/` subfolder. Moving logo.png accordingly to match the convention used by ackee, pgweb, firefox, syncthing, etc.
|
Thanks for the review @AhsanShahbaz! Fixed: moved |
… PRs Following the style used in easypanel-io#1392 (Paperclip) and easypanel-io#1396 (ZeroClaw): - Blank line between top-level sections (description, instructions, changeLog, links, contributors, schema, benefits, features, tags) - changeLog date corrected to 2026-05-13 (was an old placeholder) - Image tag pinned to v0.22.0 instead of :latest, matching the pin-to-version convention used by ZeroClaw and others - Instructions reworded to reference the UI labels (LLM Provider URL, LLM API Key) rather than internal env-var names, since users configure these via the Web UI Settings page rather than env vars
|
Sorry for the misnamed handle in my previous comment, @Ahson-Shaikh! Took another pass against #1392 (Paperclip) and #1396 (ZeroClaw) as references. Changes in the latest commit:
Happy to add a |
|
Added Ready for re-review whenever you have a moment. |
qwe-qwe — Business-Oriented AI Agent
GitHub: https://github.com/deepfounder-ai/qwe-qwe
Community: https://t.me/qwe_qwe_ai
What is qwe-qwe?
A business-oriented AI agent designed to drop into real workflows — customer ops, internal automation, knowledge retrieval, scheduled reporting, and custom integrations. Deploy on your own infrastructure and bring any OpenAI-compatible LLM provider.
Bring your own LLM: Azure OpenAI, AWS Bedrock, OpenAI, Groq, OpenRouter, DeepSeek, Together — or a local model via LM Studio / Ollama for fully on-prem deployments. Switch providers without restarting.
Features
.pyfile to extend; or create skills from chat in 60sTemplate parameters
qwe-qweghcr.io/deepfounder-ai/qwe-qwe:latesthttps://api.openai.com/v1gpt-4o-miniData persistence
All user data stored in a named volume at
/root/.qwe-qwe— conversations, memory, uploaded files, and knowledge base survive restarts.Docker image
Published to
ghcr.io/deepfounder-ai/qwe-qwe:lateston every push to main via GitHub Actions.