softengineware/bluebeam-agnt

Bluebeam Studio AI Integration (BSAI)

Overview

BSAI is a sub‑agent that connects ChatGPT‑4o with the Bluebeam Studio API and Revu scripting to automate civil take‑off and markup workflows.
The agent learns from existing, manually‑marked PDF sheets, stores this knowledge, and applies it to new drawings on demand via natural‑language commands.

Key capabilities

• 🔗 Chat interface – Issue instructions through ChatGPT Pro; the agent interprets intents and executes them.
• 🖌️ AI‑driven markup replication – Computer‑vision & NLP models analyse reference markups and reproduce them on other sheets/files.
• 💾 Persistent memory – Conversations and context are archived in mem0, Supabase SQL (relational), and Supabase Vector (embeddings) for continuous learning.
• 📚 Built‑in knowledge base – Periodically crawls Bluebeam documentation, indexes the content, and makes it queryable by the agent.
• 🛠️ Dual automation modes – Prefers Bluebeam Studio API; falls back to Revu eXtreme scripting where necessary.
• 🖧 MCP server – A FastAPI service that orchestrates sub‑components, exposes REST/WebSocket endpoints, and enables horizontal scaling.

Quick Start (Local Dev)

git clone <repo> bsai && cd bsai
cp .env.template .env   # fill in secrets
python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
uvicorn mcp.main:app --reload

Repository Layout

.
├── mcp/                 # MCP server source
├── agent/               # ChatGPT orchestration, memory, task queue
├── bluebeam/            # Bluebeam API & scripting adapters
├── knowledge/           # Docs crawler & vector ingest
├── tests/
└── docs/

Contributing

1. Fork & clone the repo
2. Create a feature branch: git checkout -b feat/<topic>
3. Run make test before opening a PR

License

MIT