PaperManager

Knowledge Features

PaperManager includes three advanced knowledge tools: an interactive graph, a multi-paper AI chat, and a direct Cypher query editor.

---

Knowledge Graph

Navigate to Graph in the nav bar for an interactive WebGL visualisation of your entire library.

Node Types

Colour	Node type
🟣 Purple	Paper
🔵 Blue	Person
🟢 Green	Topic
🟠 Orange	Tag
🩷 Pink	Project
⚫ Grey	Note

Controls

• Pan — click and drag the background
• Zoom — scroll or pinch
• Move nodes — drag individual nodes
• Click a node — opens a properties panel on the right:
  • View all node properties
  • Navigate to the paper detail page
  • Delete the node

Graph Modes

Mode	What is shown
Full graph	All node types (up to 500 nodes)
Papers only	Papers, People, Topics
Single paper	One paper with all its direct neighbours

Switch modes using the toggle in the top-left of the graph.

Display Options

Use the sliders and toggles in the control panel to adjust:

• Node size
• Link distance
• Repulsion force
• Show / hide node labels
• Show / hide edge labels

---

Knowledge Chat

Navigate to Knowledge in the nav bar to chat with Claude about multiple papers at once.

Conversation

Type a question. By default the 10 most recently added papers are used as context.

Use @mentions to bring specific papers or groups into context:

Mention syntax	What is loaded
@paper:Attention is All You Need	That specific paper
@tag:deep-learning	All papers with that tag
@topic:Protein Folding	All papers about that topic
@project:my-phd-papers	All papers in that project

Example Prompts

@tag:deep-learning What are the main architectural differences across these papers?

@topic:Protein Folding How has the approach changed from RoseTTAFold to AlphaFold?

@project:my-phd-papers Summarise the key open problems.

@paper:Attention is All You Need What positional encoding does this use?

Features

Feature	Description
SSE streaming	Answer appears token by token
Step-by-step progress	Shows which Cypher queries fetch context, how many papers are loaded
Context visualisation	Stacked bar chart of token usage per paper
Model selector	Claude Opus, Claude Work (enterprise), Ollama
Conversation history	Create new conversations, load previous ones
Compact	Summarise conversation history into a system message to free up context

Context Assembly

flowchart TD
    Q["User question"] --> P["Parse @mentions"]
    P -->|"Mentions found"| C["Cypher queries to fetch\nmatching papers"]
    P -->|"No mentions"| R["10 most recently added papers"]
    C --> A["Assemble context\n(truncated to token budget per paper)"]
    R --> A
    A --> S["Stream Claude Opus response via SSE"]

---

Cypher Editor

Navigate to Cypher in the nav bar for direct access to the Neo4j database.

Schema Browser

A live view of all:

• Node labels and their property keys
• Relationship types
• Fulltext indexes

Query Editor

Write and run raw Cypher queries. Results are shown in a table. Mutation counters show nodes created/deleted, relationships created/deleted, and properties set. Maximum 500 rows are returned.

AI Assist

Describe what you want in plain English — Ollama generates the Cypher query for you. You can then review and run it.

Example Queries

-- Papers citing a specific paper
MATCH (a:Paper)-[:CITES]->(b:Paper {title: "..."})
RETURN a.title, a.year

-- Most connected authors
MATCH (p:Person)<-[:AUTHORED_BY]-(paper:Paper)
RETURN p.name, count(paper) AS papers ORDER BY papers DESC LIMIT 10

-- Papers without summaries
MATCH (p:Paper) WHERE p.summary IS NULL RETURN p.title, p.year

-- All papers on a topic
MATCH (p:Paper)-[:ABOUT]->(t:Topic {name: "Transformers"})
RETURN p.title, p.year ORDER BY p.year DESC

-- Papers shared by a colleague
MATCH (person:Person {name: "Jan"})<-[:INVOLVES {role: "shared_by"}]-(p:Paper)
RETURN p.title, p.year

---

People

Navigate to People in the nav bar.

All people in the system are listed in the sidebar with search.

Person detail shows:

• Name and affiliation
• Papers where they are an author (AUTHORED_BY)
• Papers where they have an INVOLVES role (shared_by, working_on, collaborating, etc.)
• Research specialties (linked Topic nodes via SPECIALIZES)

People are auto-created when papers are ingested from the author list.

---

Projects

Navigate to Projects in the nav bar.

Create named collections of papers:

• Create a project with name, description, and status (active / paused / done)
• Add/remove papers from a project
• Papers can belong to multiple projects
• Filter the library by project
• Link projects as related (RELATED_TO)
• Select a project during upload, URL ingest, or bulk import to add papers automatically

---

Export & Backfill

Export

Format	How
BibTeX	GET /export/bibtex — downloads a .bib file containing all papers
JSON	Available from the Settings page

Backfill (Bulk Enrichment)

Run from the Settings page or directly via the API:

Operation	Endpoint	Description
Backfill topics	POST /backfill/topics	Run Claude Haiku topic suggestion on all papers without topics
Backfill summaries	POST /backfill/summary	Generate AI summaries for papers that have raw_text but no summary
Backfill figures	POST /backfill/figures	Extract figures from all papers that have a PDF but no figures yet

Each returns {processed, skipped, errors}.