ConsilienceTurn your files into a knowledge graph your AI reasons over.
Just write, drop in files, or code as usual — Consilience quietly builds a knowledge graph across your notes, code, design, and maps. Then one agent works on top of it: ask a complex question or hand off a whole task, and it uses everything you've built.
- Free to start
- 7 languages
- Runs on your device
Work on top of a knowledge graph, and the agent changes.
Consilience automatically weaves your workspace into an ontology knowledge graph. So the very same one-liner gets you depth — and work — that ordinary AI agents can't reach.
Every source becomes one knowledge graph.
Bring in spreadsheets, Word, PDFs, images, and URLs exactly as they are, and the key concepts and relationships are woven into a knowledge graph automatically. The agent reasons on top of it to answer and to work — a true second brain.
- Faithful conversion that never flattens into a summary
- Connected as concepts and relationships the moment you save
- No folders or tags — topic communities surface on their own
Build your knowledge graph from markdown documents.
An editor with free rein over tables, equations, diagrams, wikilinks, and a slash menu. Edit a note and the knowledge graph changes along with it.
- Real markdown, opens anywhere
- Wikilinks connect note to note
- Notes stay yours as plain originals, all the way
Code intelligence.
Understands the structure and meaning of your code to pinpoint exactly what to change — then plans, finds, fixes, and runs it to verify.
- Finds code by what it does, not just by matching text
- Traces callers and imports to see what a change affects
- CodeBridge pulls the design note behind a function into the work
A design canvas that draws what you imagine.
The agent builds visuals like slides and pages and shows them rendered, right there. Point to what you want changed and hand it to the AI.
- Rendered in place the moment it's made
- Select just the part to change and delegate to the AI
- The result is an ordinary file that opens anywhere
A map canvas that turns places into insight.
Geocoding turns the addresses and places in your graph into coordinates, and the agent authors an interactive map for each request — hotspots, by-region rollups, reachability, routes. The coordinate math is done by libraries, never guessed, so it stays exact.
- Geocoding turns addresses and places into coordinates
- Hotspots, regional rollups, reachability, and routes
- Pick a spot on the map and follow it to its graph node
One agent, everywhere
Not just notes. Code, terminal, external apps, the web, even the screen — the same agent handles it all, under the permissions you set.
Plan mode
Complex work starts with a plan; edits are blocked until you approve
Multi-file editing
Plans changes across many files and applies them in one pass
Terminal · run commands
Runs builds and tests, reads the results, and decides what's next
Direct computer control
Watches the screen and works desktop apps with mouse and keyboard
External app connections
Operates external apps like Notion directly, under your permissions
Web search
Pulls fresh information from the web for grounding when needed
Ontology search
Follows concepts and relationships to find things across the whole workspace
Topic map
Surfaces which topics your notes cluster into, as a map
Domain vocabulary
Learns the terms of your field and understands them your way
Synonym · relation suggestions
Finds matching meanings and hidden links and proposes them first
Sourced answers
Not guesses but your own sentences, down to the source note
Markdown editor
Tables, equations, and diagrams, right inside real markdown files
Multiple model choice
Pick the latest model and reasoning effort for each task yourself
Experience memory
Keeps what you've taught it and what it's confirmed as notes, and gets smarter
Git integration
Review changes and handle commits and branches right there
Permission control
Set run permissions per tool in advance, so it only asks when it must
Workspace sharing
Share a workspace as a file or link and open it as-is
Geocoding · maps
Turns the places in your graph into coordinates and maps them to analyze
A new way to work with knowledge.
Real problems teams across industries ran into, and how they solved them. Swipe through the cards to read the cases.
Just ask, and the clause comes too
Instead of paging through the Subcontracting Act themselves, construction field staff just ask in plain language. A customer PoC where Consilience answers with the supporting clause and a source link attached.
Ask the 32-page Subcontracting Act something like "if we pay more than 60 days late, what about the late-payment interest?" and get an answer with the supporting clause and its source link attached.
Customer PoC · DemoRead case →Personal knowledgeYou don't need the right search term
A product manager with five years and 1,800 notes finds his own writing using a concept that appears nowhere in the title. Not a real customer story — a hypothetical scenario we thought through.
Across 1,800 notes, asking "what did I write up about the subscription economy last year?" surfaces notes that never use the phrase in the title or the body as "Related", and the "Where this came from in your notes" card separates the direct hits from the ones reached by walking relations.
Example · Hypothetical scenarioRead case →ManufacturingThe handover, 18 months before he retires
A 30-year veteran of the paint line retires in 18 months. A hypothetical scenario for pulling 15 years of scanned handwritten maintenance logs and a weekly interview recording into one workspace.
From 15 years of scanned handwritten maintenance logs and a weekly interview recording, a question like "booth 3 pressure hunting, past cases on humid days" comes back with what was done at the time and links to the log notes it came from.
Example · Hypothetical scenarioRead case →Pharma · Chemical R&DThree lines of evidence, one channel
A pharma and chemical R&D case where internal reports, patents, and external papers are queried from one channel, and every answer comes with its source (document name, page, metadata). At the customer PoC and demo stage, with quantitative performance still under validation.
Ask internal technical reports, patents, and external papers from one channel without moving windows, and get a report where the document name and page, the patent conditions, and the paper metadata are attached as sources.
Customer PoC · DemoRead case →B2B startupWhy did we build it this way?
The CTO of a 30-person B2B SaaS company questions 200-odd meeting notes and a pile of architecture decision records one decision at a time. A hypothetical scenario about a workspace where the original sentence and its revision trail outlast the person who wrote it.
Ask for "every decision we made about multi-tenancy, newest first" and six decisions come back with who said each one, an "Architecture sync (2021-11).md · line 14" citation, and a "Superseded" badge on the two that got reversed.
Example · Hypothetical scenarioRead case →Investigative journalismFinding the links in 4,200 scans
Leave 4,200 leaked document scans in a folder on your laptop and ask, in one line, how the companies, the people, and the addresses are tangled together. A hypothetical scenario for a local newsroom's investigative desk.
With 4,200 leaked scans sitting in a folder on the laptop, ask in one line for "companies registered at a building owned by the procurement director's spouse, and the contracts those companies appear in" — and get an answer that quotes the sentence from the corporate registry.
Example · Hypothetical scenarioRead case →University Academic AffairsAsking the rulebook on one screen
A university academic affairs office that used to hunt for the governing clause in a vast rulebook tied asking, loading, and editing into one workspace. Every answer comes with the source text of the clause. Having passed security review, it's near deployment.
Ask in natural language which clause governs a matter like a major change or a leave of absence and get the regulation's name, the clause's location, and the clause's source text on one screen; upload a revised PDF and it lands in the index automatically; edit the regulation right in the system's editor.
Near deploymentRead case →Grad school · ImmunologyWhy did I switch protocols again?
A fourth-year immunology PhD student, four months out from her defense, gets her own two-year-old decision back verbatim. Not a real customer story — a hypothetical scenario we thought through.
In a workspace holding 900 literature notes and 400 experiment records, asking "every judgment I wrote about Protocol B, in order" returns the sentence written two years ago word for word, along with which note it came from and which line.
Example · Hypothetical scenarioRead case →City emergency opsOne bridge, two reports that disagree
The emergency operations center of a city of 400,000. When a district's HWP report and a fire radio log say different things about the same bridge, you find those two lines side by side before the 21:00 briefing. This is a hypothetical scenario, not a real customer story.
From a folder mixing district HWP reports, fire radio logs, and spreadsheets, ask "every report on Dongbu Bridge, in time order" and get an answer with "Dongbu district situation report #3 · line 12" attached — without sending a single document outside the government network. The two reports that disagree come back stacked in the answer, and nothing lands until a person confirms it.
Example · Hypothetical scenarioRead case →Global ManufacturingScattered close, in one place
Putting the data of six affiliates — scattered across SAP, Excel, and email — onto a single ontology, from automatic reconciliation all the way to consolidation adjustments. The story of a global manufacturer's K-IFRS consolidated-close PoC.
Map six affiliates' SAP extracts, Excel files, and emailed attachments onto a common ontology, classify intercompany reconciliation differences by cause — timing, amount, classification — and get recommended consolidation adjustment entries out as Excel.
Customer PoC · DemoRead case →Game studioCatch lore contradictions before fans do
The narrative director at a 15-person game studio has 700-odd lore documents and one question: what year did the mentor die? Not a real customer story — a hypothetical scenario we worked through.
Load 700-odd lore docs, quest scripts, and character sheets, ask "what do the documents say about the year the mentor died?", and the character sheet that says RE 412 comes back alongside the chapter 7 script that puts him at the RE 414 Harvest Festival — each with its verbatim sentence and note name attached.
Example · Hypothetical scenarioRead case →Personal investingNews that breaks your assumption, pinged
A retail investor with eight years of research notes pulls the assumption he wrote at buy time back up, sets a watch on the words that assumption hangs on, and gets a notification when new stories come in. Not a real customer — a hypothetical scenario we thought through.
The buy-thesis note and the incoming news note attach to the same ticker entity, a matching item fires a "Watch: capacity expansion" notification, and asking about that holding returns an answer that quotes sentences from both notes together.
Example · Hypothetical scenarioRead case →Questions, answered.
Is my data private?
Yes. Consilience is local-first — your notes and the knowledge graph live on your device. We don't train any model on your notes, indexing and search run on-device, and web access is off by default. The agent asks permission before it acts.
Does it work offline?
The app, your notes, and your knowledge graph all work offline. The agent needs a connection only to reach a language model; everything you've built stays and reads on your device.
What can I bring in?
Notes, PDFs, Word docs, spreadsheets, images, web pages, and code — imported faithfully as their full content, never flattened into a summary.
Can I move over from Obsidian or Notion?
Yes. Bring your Markdown and files as they are. Your notes stay plain, standard files you can open anywhere and export anytime — no lock-in.
Which platforms are supported?
macOS and Windows today. It's free to start, and downloads the right build for your system automatically.
What do paid plans change?
Only how much you can use the agent. Every plan — Free included — is the complete workspace; paid tiers simply raise your agent usage. Cancel anytime.