This repository is the concrete demo project for running SuperOptiX agents against SurrealDB-backed memory, RAG, and GraphRAG workflows. It is intended to be the companion implementation for the SuperOptiX + SurrealDB integration story, with generated playbooks, compiled pipelines, and recorded traces checked into the repo on purpose.
This repo is meant to be a working demonstration, not a minimal starter template. It shows how SuperOptiX-generated agents can share the same SurrealDB-backed retrieval and memory layer across multiple frameworks while keeping the generated artifacts visible and inspectable.
Run the DSPy RAG demo and verify that retrieval is grounded on the seeded token NEON-FOX-742.
pip install -U "superoptix[frameworks-dspy]"
python -m superoptix.agents.demo.setup_surrealdb_seed
super agent pull rag_surrealdb_dspy_demo
super agent compile rag_surrealdb_dspy_demo --framework dspy
super agent run rag_surrealdb_dspy_demo --framework dspy --goal "What is NEON-FOX-742?"Expected result:
- the final answer should mention
NEON-FOX-742 - the run should complete successfully
- traces should be written to
.superoptix/traces/
For the broader multi-framework walkthrough, use the published SuperOptiX guide:
For the main SuperOptiX website, see:
- SuperOptiX-generated agents using a shared SurrealDB backend
- Grounded RAG checks using the seeded retrieval token
NEON-FOX-742 - GraphRAG examples on the same SurrealDB data model
- Multiple framework adapters compiled from SuperOptiX playbooks
- Checked-in execution traces under
.superoptix/traces/
| Framework | Basic RAG Demo | GraphRAG Demo |
|---|---|---|
| DSPy | rag_surrealdb_dspy_demo |
graphrag_surrealdb_dspy_demo |
| OpenAI | rag_surrealdb_openai_demo |
present in SuperOptiX docs |
| Claude SDK | rag_surrealdb_claude_sdk_demo |
present in SuperOptiX docs |
| PydanticAI | rag_surrealdb_pydanticai_demo |
graphrag_surrealdb_pydanticai_demo |
| CrewAI | rag_surrealdb_crewai_demo |
present in SuperOptiX docs |
| Google ADK | rag_surrealdb_adk_demo |
present in SuperOptiX docs |
| DeepAgents | rag_surrealdb_deepagents_demo |
present in SuperOptiX docs |
The generated pipeline files are committed intentionally so this repository can be inspected as a working SuperOptiX output project, not just as source playbooks.
Framework-specific walkthrough links:
- DSPy: https://superagenticai.github.io/superoptix/examples/agents/surrealdb-frameworks-demo/#__tabbed_1_1
- OpenAI: https://superagenticai.github.io/superoptix/examples/agents/surrealdb-frameworks-demo/#__tabbed_1_2
- Claude SDK: https://superagenticai.github.io/superoptix/examples/agents/surrealdb-frameworks-demo/#__tabbed_1_3
- Microsoft: https://superagenticai.github.io/superoptix/examples/agents/surrealdb-frameworks-demo/#__tabbed_1_4
- PydanticAI: https://superagenticai.github.io/superoptix/examples/agents/surrealdb-frameworks-demo/#__tabbed_1_5
- CrewAI: https://superagenticai.github.io/superoptix/examples/agents/surrealdb-frameworks-demo/#__tabbed_1_6
- Google ADK: https://superagenticai.github.io/superoptix/examples/agents/surrealdb-frameworks-demo/#__tabbed_1_7
- DeepAgents: https://superagenticai.github.io/superoptix/examples/agents/surrealdb-frameworks-demo/#__tabbed_1_8
SuperOptiX traces are stored in:
.superoptix/traces/
Current trace files include framework-specific runs such as:
.superoptix/traces/rag_surrealdb_dspy_demo.jsonl.superoptix/traces/rag_surrealdb_openai_demo.jsonl.superoptix/traces/rag_surrealdb_adk_demo.jsonl.superoptix/traces/rag_surrealdb_deepagents_demo.jsonl
These traces are part of the demo value of the repo and show actual SuperOptiX run activity over the SurrealDB examples.
Architectural background:
Framework-by-framework guide:
SuperOptiX website:
Live repository:
- generated playbooks
- compiled pipelines
- compiled spec JSON files
.superoptix/traces/output from demo runs
This repository keeps the standard SuperOptiX project layout intact.
surrealoptix/
├── surrealoptix/
│ ├── agents/
│ │ ├── rag_surrealdb_*_demo/
│ │ └── graphrag_surrealdb_*_demo/
│ ├── knowledge/
│ ├── memory/
│ ├── tools/
│ ├── guardrails/
│ ├── protocols/
│ ├── evals/
│ ├── optimizers/
│ ├── teams/
│ └── servers/
├── .super
├── .superoptix/
│ └── traces/
├── pyproject.toml
├── README.md
└── tests/
- The published SuperOptiX docs page is the detailed operator guide for framework-specific commands and troubleshooting.
- This repo is the implementation companion: playbooks, generated pipelines, and traces.
- Start with the DSPy path in this README, then use the published docs for the broader multi-framework walkthrough.
- If a run completes but does not mention
NEON-FOX-742, reseed SurrealDB and verify the agent is pointing at the expected demo database.
pytest tests/MIT