sparkit
SPARKIT-science/sparkit-mcp
Ask SPARKIT, a scientific research agent, questions and get cited Markdown reports based on literature search and paper reading.
- Cloud integrations
Installation, supported tools and required credentials are maintained by the project.
Project details
- GitHub stars
- 1
- Language
- Python
- Forks
- 0
- Latest release
- v0.1.05mo ago
- Activity
- updated 4mo ago
First MCP server?Start with the basics
Check for existing competition before building: scan GitHub, Hacker News, npm, PyPI and Product Hunt for similar projects, duplicate likelihood and pivot hints.
ITASCA PFC discrete element simulation: browse documentation, execute scripts, capture plots and manage long-running tasks via a bridge to the PFC GUI.
Deep research via the OpenAI Responses API or Open Deep Research (smolagents), with web search and code interpreter support.
Checks every reference of a LaTeX paper against Crossref, dblp, arXiv, DataCite, PubMed and OpenAlex records: whether each cited work exists, matches its metadata and was not retracted, with no LLM in the verdict.
Self-hosted perception server: local bilingual ASR, vision, cross-lingual search, credentialed login-walled sources and a persistent relation graph.
Natural-language patent and scientific literature search via Patsnap: semantic and BM25 search, structured filters and record retrieval.
Research workspace for Zotero, Obsidian and NotebookLM: search arXiv, Semantic Scholar, PubMed and Crossref, ingest into Zotero, sync notes and verify briefs.
Zotero integration: hybrid search, PDF passages with page locators, CSL bibliographies, add by DOI, group libraries and writes including PDF annotations.