AI & Intelligence · Projects
One case, fully organized for AI — every artifact in one governed context, with work products that prove their own freshness.
Included with EdgeLex
📁 Civil — Hale v. Northstar
Project chats use the instructions, files, and saved AI work attached here.
Living project work products
📇 Matter Source Inventory CURRENT
Built Aug 14 · 419 sources
✓ deterministic build · build receipt recorded
✓ 419 source fingerprints verified
🧮 Discovery Matrix FRESH
Built Aug 14 · 19 sources
✓ deterministic build · build receipt recorded
✓ 19 current source fingerprints verified
🗺 Matter Evidence Map SCOPE NEEDED
Tests lawyer-defined propositions against the exact current Source Inventory.
Current inventory ready · lawyer proposition required
🕰 Case Chronology NOT BUILT
What happened at each stage: filings, responses, hearing outcomes, next expected hearings.
⚠ Issue Register NOT BUILT
Issues, evidence, adverse facts, missing proof, risk, and next action.
＋ Describe any work product GOVERNED
Create a governed living Project artifact from a sentence.
The problem
AI is only as good as the context it holds — and most tools make the lawyer rebuild that context in every conversation, with no way to know whether yesterday's analysis still matches today's record.
One case, engineered for AI
Created. Organized. Preflighted. Delivered.
Follow one Project from a memory policy to a deliverable in the record — every screen below is the real product.
01 · Created
A Project starts with a memory policy.
Bind a Project to a matter and its folders mirror the matter's own taxonomy. Then decide — with checkboxes, not defaults — what Lex may remember and must confirm: matter memory, document memory, related-matter context, personalization.
Use projects to organize files, chats, instructions, and memory policy for ongoing work.
When a matter is selected, project file folders mirror that matter’s Documents tab.
Memory policy
☑ Use global personalization
☑ Use active matter memory
☑ Use selected document memory
☐ Allow related matter context
☑ Confirm matter facts
☑ Confirm case events
02 · Organized & built
The case brain, with receipts.
Instructions, files, and saved AI work in one governed context — and living work products that prove their own state: FRESH with a recorded build receipt and every source fingerprint verified, SCOPE NEEDED when a lawyer proposition is missing, NOT BUILT when nothing pretends otherwise. Describe any work product in a sentence and it becomes a governed living artifact.
How new authority arrives →📁 Civil — Hale v. Northstar
Project chats use the instructions, files, and saved AI work attached here.
Living project work products
📇 Matter Source Inventory CURRENT
Built Aug 14 · 419 sources
✓ deterministic build · build receipt recorded
✓ 419 source fingerprints verified
🧮 Discovery Matrix FRESH
Built Aug 14 · 19 sources
✓ deterministic build · build receipt recorded
✓ 19 current source fingerprints verified
🗺 Matter Evidence Map SCOPE NEEDED
Tests lawyer-defined propositions against the exact current Source Inventory.
Current inventory ready · lawyer proposition required
🕰 Case Chronology NOT BUILT
What happened at each stage: filings, responses, hearing outcomes, next expected hearings.
⚠ Issue Register NOT BUILT
Issues, evidence, adverse facts, missing proof, risk, and next action.
＋ Describe any work product GOVERNED
Create a governed living Project artifact from a sentence.
03 · Preflighted
Formal drafting starts by freezing the request.
The preflight records scope, sources, model budget, and fixed tool authority — and creates nothing: no task, no run, no model call, no document. Governance happens before the first token, not after the mistake.
Saving freezes this request. It does not start Lex or create a document.
04 · Delivered
Deliverables land in the record, not in a chat window.
A finished draft arrives as a normal DMS version with AI provenance — 'Verified deliverable from legal graph' — ready for EdgeDocs or Word review before anything external happens.
The document's life from here →🗓 This Month (2 versions) · Group by Date
Verified deliverable from legal graph msa_obligation_register.v1 — every obligation traced to its clause span.
👤 Maya Chen (Attorney) · 301 words · 10 KB
Initial version.
👤 Maya Chen (Attorney) · 294 words · 10 KB
The difference
Context is engineered, not accumulated.
Most AI tools let context pile up until nobody knows what the model is looking at. A Project makes it explicit: these sources, these instructions, this memory policy — and work products that can prove they were built from the current record, not last month’s.
How Lex uses it →Build receipts
Every living work product records how and when it was built — deterministically.
Source fingerprints
419 sources verified against the current record. Stale is shown, never hidden.
Memory by policy
What Lex remembers and must confirm is a per-Project decision, not a default.
Capabilities
What it does
One governed context per case
A Project binds instructions, files, saved AI work, and memory policy to a matter. Project chats start from the whole case, not a blank box.
Living work products
Discovery matrices, chronologies, issue registers, evidence maps — built deterministically, stamped with a build receipt, and verified against current source fingerprints. FRESH means provably fresh; stale is shown, never hidden.
Immutable preflights
Formal drafting starts by freezing the request: scope, sources, model budget, and fixed tool authority — recorded before a single model call. The preflight itself creates nothing and spends nothing.
Deliverables land in the record
A draft arrives as a normal DMS version with AI provenance, ready for EdgeDocs or Word review — never a loose file in a chat window.
Memory under policy
Each Project chooses what Lex may remember and confirm: matter memory, document memory, related-matter context, personalization — checkboxes, not defaults.
In your control
The Project is a governance boundary: its instructions, sources, and memory policy are explicit and inspectable, its work products carry receipts, and nothing it produces enters the record without review.
Related
Works with
Lex AI & Agents
Lex isn't one application among many. It's the horizontal layer connecting matters, documents, email, messages, meetings, deadlines, and billing — grounded in your data, gated by your approvals.
Lex Delegations
Standing delegations you create in plain language. Lex watches the matter, wakes when your firm's events fire, does the work inside your approval gates — and every run lands on the task ledger like any other member of the team.
Matters
Matter lifecycle, conflict checks, risk scoring, related-proceeding tracking, and a matter workspace the AI understands.
See EdgeLex on your own terms.
We'll walk through self-hosting, model control, and your firm's workflows.