vigmarcarlo/OntoPrune

Neuro-Symbolic Context Pruning Middleware for Local SLMs and Cloud LLMs
English | Español | 📄 Technical Whitepaper (PDF) | 📄 Whitepaper en Español (PDF)
OntoPrune is an ultra-lightweight ( Ret: ... |
| Flutter / Dart | .dart | tree-sitter-dart | abstract class ... { Ret method(); } |
| Java / Spring Boot | .java | tree-sitter-java | public interface ... { Ret method(); } |
| TypeScript / React | .ts, .tsx, .js | tree-sitter-typescript | export interface ... { method(): Ret; } |
## 🛠️ Usage Modes
### 1. Native Model Context Protocol (MCP) Server
OntoPrune runs out of the box as an MCP server (ontoprune-mcp) compatible with Claude Desktop, Cursor, Gemini CLI, or Antigravity IDE:
bash
ontoprune-mcp
Configuration in claudedesktopconfig.json:
json
"mcpServers": {
"ontoprune": {
"command": "ontoprune-mcp"
Exposed MCP Tools:
prunecontext(filepath, targetsymbol, format='stubs'): Extracts the minimal dependency contract resolving cross-file imports.
verifyresponse(responsecode, contractorfile): Deterministically validates generated code against authorized contracts.
### 2. Command Line Interface (CLI)
bash
# Prune a target method across multi-file projects:
ontoprune translate src/services/OrderService.java processOrder --format stubs
# Direct streaming pipeline with local Ollama:
ontoprune translate services/orderservice.py procesarorden | ollama run qwen2.5-coder:3b
# Deterministically verify LLM output against contract:
ontoprune check --file generatedsolution.py --contract contract.py
### 3. Python API
python
import ontoprune
# 1. Prune a multi-module project to a minimal typed contract
context = ontoprune.translate(
"services/orderservice.py",
target="procesarorden",
fmt="stubs",
multimodule=True,
print(context)
# 2. Verify model output against the contract
violations = ontoprune.check(llmcoderesponse, against=context)
if not violations:
print("Code is 100% compliant and free of hallucinations!")
## 🧪 Test Suite
bash
pytest tests/
# 46 passed in 1.40s
## 🔬 Reproduce Empirical Benchmarks
You can independently replicate the benchmark evaluation in isolated sandboxes on your machine:
bash
# 1. Quick deterministic validation in sandbox (no API key needed):
ontoprune benchmark --backend mock --analyze
# or via Make:
make benchmark-mock
# 2. Live evaluation against Google Gemini (requires GEMINIAPIKEY):