Skip to content

Releases: temiroff/Blacknode

Blacknode v0.1.0 Preview - visual AI workflows with MCP control

21 May 15:03
@temiroff temiroff
v0.1.0-preview
3bb7042

Choose a tag to compare

View all tags
Blacknode v0.1.0 Preview - visual AI workflows with MCP control Pre-release
Pre-release

Watch the Blacknode MCP + NVIDIA NIM preview demo
Run workflow

Summary

Blacknode is a local-first visual workflow builder for AI agents. This preview includes a React node editor, Python runtime, portable workflow JSON, CLI execution, Python export, reusable templates, run history, and an MCP server so agents can create and operate workflows through typed tools.

Highlights

  • Visual node editor with typed, color-coded ports.
  • Python graph runtime with per-run evaluation.
  • Workflow CLI: validate, run, export Python, and launch MCP.
  • MCP server for AI-agent workflow building and editor control.
  • Templates for LLM chat, NVIDIA NIM, research flows, Python tools, tool agents, and subnets.
  • Run history with event timelines, model/tool call visibility, result previews, and error inspection.
  • Local provider-key handling for Anthropic, OpenAI, NVIDIA NIM, and local/Ollama-style models.
  • CI covering Python tests, editor build, and Rust workspace check.

Try It

No API key required:

pip install -e .
python -m blacknode.cli validate templates\text-pipeline.json
python -m blacknode.cli run templates\text-pipeline.json
python -m blacknode.cli export-python templates\subnet-tool-call.json --output subnet_tool_call.py
python subnet_tool_call.py

Start the editor on Windows:

start.bat

Start the MCP server:

pip install -e ".[mcp]"
blacknode mcp

Then use the prompts in docs/quickstart-mcp.md or docs/mcp-test-prompts.md.

Validation

Current local validation:

python -m pytest
npm run build
cargo check
python scripts\smoke_test_mcp.py

Expected status:

  • Python tests pass.
  • Editor production build passes.
  • Rust workspace check passes.
  • MCP smoke test reports a healthy server.

Known Limitations

  • Public preview: graph APIs and workflow internals may change.
  • Windows has the smoothest launch path through start.bat; macOS/Linux users should use the manual two-terminal start.
  • Arbitrary Python tool execution is a developer feature and should not be treated as a hardened sandbox.
  • NVIDIA NIM, OpenAI, and Anthropic demos require provider credentials.
  • Rust crates are present as a future runtime direction, but the packaged Python runtime is currently canonical.
  • No hosted cloud workspace or team collaboration yet.
Loading