Repolode
Submit
shuxiachai

shuxiachai/academic-commercialization-agent

Multi-agent LLM pipeline that turns a research paper or topic into a cited scorecard on technology readiness, patents, market and risks.

View on GitHubWebsite
Stars
809
Forks
98
Open issues
12
Last commit
7 hours ago

What is it?

This project takes a research paper (PDF) or a topic and produces a commercialization triage report. Deterministic code first retrieves and validates sources into a frozen registry; six LLM stages then reason only over those sources. The scorecard covers TRL, MRL, patent strength, market accessibility and evidence confidence, and the weighted total is computed by code rather than by the model.

The pipeline is built with CrewAI and FastAPI, with a browser client that needs no build step, a CLI, Docker and Railway deployment. Runs are checkpointed, so an interrupted run can resume as a new linked run. Reports can be exported as Markdown or PDF, and optional OpenTelemetry tracing is supported.

The project publishes its benchmark protocols and results, including held-out checks and limitations. It is meant for research triage, not as technical, legal, or investment due diligence.

Key features

  • 01Turns a paper PDF or research topic into a scored commercialization assessment.
  • 02Scores TRL, MRL, patent strength, market accessibility and evidence confidence.
  • 03Every claim is tied to registered source IDs, with reliability warnings.
  • 04Interrupted runs resume from validated checkpoints as new linked runs.
  • 05Exports reports as Markdown or PDF; supports optional OpenTelemetry tracing.
  • 06Works with Qwen, DeepSeek, Anthropic or OpenAI models and Tavily or Serper search.

Use cases

  • Triaging university research for licensing or technology-transfer offices.
  • Getting a first, cited view of readiness and patent landscape for a emerging technology.
  • Studying how a multi-agent LLM pipeline is structured, evaluated and benchmarked.

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/shuxiachai/academic-commercialization-agent.git
  2. 2

    Enter the project directory

    cd academic-commercialization-agent
  3. 3

    Install dependencies with uv (Python 3.11 or 3.12)

    uv sync
  4. 4

    Copy .env.example to .env and add your own LLM and search API keys, then start the server

    uv run uvicorn api.main:app --reload

Things to know

  • Real analysis incurs paid LLM and search provider usage that you must fund with your own keys.
  • Scores are coarse signals; the README reports TRL under-scoring of mature technologies in a held-out check.
  • Reports support research triage only, not technical, legal, investment or freedom-to-operate due diligence.

In collections

Open-source AI agent frameworks & apps

Topics

#ai-agent#multi-agent#llm#crewai#fastapi#rag#research-tools#technology-transfer#academic-research#ai-agents

FAQ

What is academic-commercialization-agent?
Multi-agent LLM pipeline that turns a research paper or topic into a cited scorecard on technology readiness, patents, market and risks. This project takes a research paper (PDF) or a topic and produces a commercialization triage report. Deterministic code first retrieves and validates sources into a frozen registry; six LLM stages then reason only over those sources. The scorecard covers TRL, MRL, patent strength, market accessibility and evidence confidence, and the weighted total is computed by code rather than by the model.
Is academic-commercialization-agent free to use? What license does it have?
academic-commercialization-agent is open source under the MIT license. Check the license terms before using it in commercial products.
Is academic-commercialization-agent still maintained?
Actively maintained. The last commit to academic-commercialization-agent was on Nov 7, 2026.
What can I use academic-commercialization-agent for?
Triaging university research for licensing or technology-transfer offices. Getting a first, cited view of readiness and patent landscape for a emerging technology. Studying how a multi-agent LLM pipeline is structured, evaluated and benchmarked.
How do I install academic-commercialization-agent?
Clone the repository: git clone https://github.com/shuxiachai/academic-commercialization-agent.git Enter the project directory: cd academic-commercialization-agent Install dependencies with uv (Python 3.11 or 3.12): uv sync Copy .env.example to .env and add your own LLM and search API keys, then start the server: uv run uvicorn api.main:app --reload
Where can I get academic-commercialization-agent?
The source code and installation instructions are on GitHub: https://github.com/shuxiachai/academic-commercialization-agent

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects