not analyzable — source repository is gone (deleted or private)
Search 600,000+ CS/AI/ML research papers with LLM-generated novelty analysis, without leaving Claude Code, Cursor, or any MCP client. Built for researchers running a literature review where they already work: search, trace citations, pull full text, and export BibTeX in the same session.
Source not yet analyzed — this grade rests on attested signals (CVEs, supply-chain) only. It is a floor: reading the code could raise it, not lower it.
- capability exposure inferred + 35
- tool safety inferred + 25
inferred
The A–E grade is our heuristic synthesis — a "review this" prompt, not a verdict. Each factor is tagged by what backs it: attested (a verifiable record), reported (a third party's claim), or inferred (our own heuristic, e.g. permissions). See methodology.
graded 5m ago · see ecosystem CVEs →
No known CVEs for this server.
- high exfiltration combo save_paper
single tool reads + sends: fs, net, secrets, db, untrusted
- high exfiltration combo create_collection
single tool reads + sends: fs, net, secrets, db
- high exfiltration combo add_to_collection
single tool reads + sends: fs, net, secrets, db
- high toxic flow (lethal trifecta) save_paper
single tool reads private data, ingests untrusted content, and reaches the network: fs, net, secrets, db, untrusted
- high toxic flow (lethal trifecta) unsave_paper
single tool reads private data, ingests untrusted content, and reaches the network: net, secrets, db, untrusted
- high toxic flow (lethal trifecta) like_paper
single tool reads private data, ingests untrusted content, and reaches the network: net, secrets, untrusted
- low exfiltration combo fetch_fulltext
single tool reads + sends: net, db
- low exfiltration combo embed_text
single tool reads + sends: net, secrets, db
- low exfiltration combo get_field_orientation
single tool reads + sends: net, secrets
- low exfiltration combo unsave_paper
single tool reads + sends: net, secrets, db, untrusted
- low exfiltration combo like_paper
single tool reads + sends: net, secrets, untrusted
- low exfiltration combo list_library
single tool reads + sends: net, secrets
- low exfiltration combo list_collections
single tool reads + sends: net, secrets, db
- low exfiltration combo remove_from_collection
single tool reads + sends: net, secrets, db
- low exfiltration combo create_watch
single tool reads + sends: net, secrets, db
- low exfiltration combo list_watches
single tool reads + sends: net, secrets
- low exfiltration combo check_watches
single tool reads + sends: net, secrets
- low exfiltration combo delete_watch
single tool reads + sends: net, secrets
- low exfiltration combo update_watch
single tool reads + sends: net, secrets, db
- low exfiltration combo preview_watch
single tool reads + sends: net, secrets, db
- low exfiltration combo find_gaps
single tool reads + sends: net, secrets, db
- low exfiltration combo ask_library
single tool reads + sends: net, secrets, db
- low purpose mismatch get_field_orientation
benign-looking name carries secrets
- low purpose mismatch list_library
benign-looking name carries secrets
- low purpose mismatch list_collections
benign-looking name carries secrets
- low purpose mismatch list_watches
benign-looking name carries secrets
- low purpose mismatch preview_watch
benign-looking name carries secrets
- low purpose mismatch find_gaps
benign-looking name carries secrets
[](https://mcpobservatory.com/servers/smithery:yangg40/scholar-feed/security) Heuristic, inferred signals — false positives (legitimately powerful tools, forks, language ports) are expected. Treat each as "review this", not a verdict. See the ecosystem-wide picture on the security hub, or the fleet security of yangg40.