arscontexta 0.6.0

package/methodology/legal case management uses precedent chains with regulatory change propagation.md

@@ -0,0 +1,892 @@
+---
+description: Legal case management knowledge system — inspirational composition showing derived architecture for precedent chain tracking, regulatory change propagation, and cross-matter intelligence
+kind: example
+domain: legal
+topics: ["[[domain-compositions]]"]
+---
+
+# legal case management uses precedent chains with regulatory change propagation
+
+A derived cognitive architecture for a litigation attorney who needs to stop relying on memory for precedent validity, regulatory changes, and cross-matter patterns. Built kernel-up from the 14 universal primitives, adapted through the natural vocabulary of legal practice, and optimized for the one thing agents do that attorneys cannot: hold every case brief, every regulatory update, every argument ever drafted in working memory simultaneously, ensuring that no relied-upon precedent goes stale and no relevant prior work product goes undiscovered.
+
+This is the most schema-heavy domain in the example catalog. Legal practice generates dense, structured metadata — case citations, jurisdictional hierarchies, regulatory frameworks, procedural timelines, evidentiary chains. The YAML frontmatter is not an overhead cost here. It is the mechanism that transforms a filing cabinet of PDFs into a queryable legal knowledge graph.
+
+---
+
+## Persona
+
+**Diana**, 38, senior associate at a mid-size litigation firm specializing in technology and data privacy law. She handles 12-15 active matters across multiple jurisdictions (federal, California, New York, EU/GDPR), involving contract disputes, regulatory compliance, IP licensing, and data breach response. She has seven years of briefs, research memos, deposition summaries, and strategy documents spread across a document management system (DMS) that is searchable by keyword but has no conceptual awareness.
+
+Diana's pain points are specific:
+1. She cited a case in a motion last month that opposing counsel showed had been distinguished by a recent appellate decision. She did not know because her DMS does not track precedent chains.
+2. A GDPR amendment affected three active matters, but she only updated two because she did not realize the third was implicated until the client called.
+3. She rewrote an argument from scratch for a new matter that she had already developed (better) in a brief two years ago, because search for "trade secret misappropriation damages" returned 200 results and she gave up scrolling.
+
+Diana does not need better document management. She needs a knowledge graph where precedents link to their treatment history, regulations link to every matter they affect, and arguments link to the evidence that supports them — and an agent that monitors all of these relationships continuously so that no change goes unnoticed.
+
+---
+
+## Configuration
+
+The 8 dimensions derived for litigation case management:
+
+| Dimension | Position | Rationale |
+|-----------|----------|-----------|
+| Granularity | Atomic for legal analysis, compound for case files | Each legal principle, case holding, and argument is a distinct, reusable unit. But a case file (matter note) is inherently compound — it tracks parties, deadlines, strategy, and related matters together because they are never accessed independently. |
+| Organization | Flat with wiki links + jurisdictional tagging | Legal knowledge is cross-jurisdictional by nature. A privacy principle applies in California and GDPR. A contract interpretation applies across multiple matters. Folders by jurisdiction would duplicate notes. Flat structure with jurisdictional metadata enables cross-jurisdictional queries. |
+| Linking | Explicit + typed + directional | Legal relationships are inherently typed: precedent → holding (establishes), case → precedent (cites/relies on), case → precedent (distinguishes), regulation → matter (applies to), evidence → argument (supports). Generic "related" links would lose the legal semantics that make the graph useful. |
+| Metadata | Dense — the densest in the example catalog | Every legal artifact carries structured metadata: citations, jurisdictions, courts, dates, parties, procedural postures, standards of review. This metadata is not optional decoration — it is how legal knowledge is queried. "Find all contract disputes in the 9th Circuit where the court applied a heightened pleading standard" is a YAML query, not a keyword search. |
+| Processing | Heavy | Legal research requires the full pipeline: case briefing (extract holdings and reasoning), precedent chain mapping (trace how cases have been treated), argument construction (link evidence to legal elements), and compliance verification (check all regulations against all matters). Each phase produces different analytical artifacts. |
+| Formalization | High from day one | Legal work is inherently formal. Citations have required formats. Courts have hierarchies. Statutes have numbering systems. The schema should mirror the formality of the domain from the start, not evolve toward it. |
+| Review | Weekly for active matters, immediate for regulatory changes, quarterly for precedent chain validation | Active matters require weekly deadline and strategy review. Regulatory changes require immediate impact assessment — a new ruling cannot wait for the weekly review. Precedent chains should be validated quarterly to catch cases that have been overruled or distinguished. |
+| Scope | Individual practitioner with firm-wide potential | Diana's system starts as her personal practice intelligence. But the schema is designed so that multiple attorneys could share the same precedent graph, with matter-specific notes remaining private and legal principle notes being shared. |
+
+---
+
+## Vault Structure
+
+```
+vault/
+├── 00_inbox/                    # raw capture
+│   ├── cases/                   # new case law to brief
+│   ├── regulations/             # regulatory updates to process
+│   ├── contracts/               # agreements to review
+│   └── research/                # legal articles, CLE materials
+├── 01_thinking/                 # active legal knowledge (flat)
+│   ├── index.md                 # hub MOC
+│   ├── precedents.md            # domain MOC: case law principles
+│   ├── regulatory.md            # domain MOC: regulatory landscape
+│   ├── contracts.md             # topic MOC: contract law patterns
+│   ├── privacy-law.md           # topic MOC: privacy and data protection
+│   ├── ip-licensing.md          # topic MOC: IP and licensing
+│   ├── trade-secrets.md         # topic MOC: trade secret law
+│   ├── evidence-standards.md    # topic MOC: evidentiary standards
+│   ├── procedural.md            # topic MOC: procedural rules
+│   │
+│   ├── transformative-use-test-requires-new-expression-not-new-purpose.md
+│   ├── CCPA-private-right-of-action-limited-to-data-breaches.md
+│   ├── inevitable-disclosure-doctrine-disfavored-in-ninth-circuit.md
+│   ├── heightened-pleading-applies-to-fraud-based-trade-secret-claims.md
+│   ├── GDPR-article-17-right-to-erasure-overridden-by-legal-hold.md
+│   ├── contractual-limitation-of-liability-unenforceable-for-willful-breach.md
+│   └── ...
+├── 02_archive/                  # closed matters, retired precedents
+│   ├── matters/                 # completed engagements
+│   │   ├── 2025-chen-v-dataflow/
+│   │   └── 2024-acme-license-dispute/
+│   └── precedents/              # overruled or superseded case law
+├── 03_matters/                  # active client matters (the engine)
+│   ├── chen-v-dataflow.md       # matter file
+│   ├── techcorp-gdpr-compliance.md
+│   ├── synthex-trade-secret.md
+│   ├── meridian-license-dispute.md
+│   └── ...
+├── 04_legal/                    # legal work product organized by type
+│   ├── briefs/                  # argument outlines and filed briefs
+│   │   ├── chen-mtd-opposition.md
+│   │   ├── synthex-pi-motion.md
+│   │   └── ...
+│   ├── research-memos/          # legal research memoranda
+│   │   ├── trade-secret-damages-ninth-circuit.md
+│   │   ├── gdpr-data-breach-notification-timeline.md
+│   │   └── ...
+│   ├── depositions/             # deposition summaries
+│   │   ├── chen-depo-wong-2026-01-28.md
+│   │   └── ...
+│   └── evidence-maps/           # evidence-to-argument linking
+│       ├── synthex-evidence-map.md
+│       └── ...
+├── 05_meta/                     # infrastructure
+│   ├── templates/
+│   │   ├── case-brief.md
+│   │   ├── matter.md
+│   │   ├── research-memo.md
+│   │   ├── regulatory-update.md
+│   │   ├── precedent-principle.md
+│   │   ├── argument-outline.md
+│   │   ├── deposition-summary.md
+│   │   ├── evidence-map.md
+│   │   ├── contract-review.md
+│   │   └── compliance-checklist.md
+│   ├── logs/
+│   │   ├── observations.md
+│   │   ├── observations/
+│   │   ├── tensions.md
+│   │   └── tensions/
+│   ├── scripts/
+│   │   ├── precedent-check.sh    # validate cited cases still good law
+│   │   ├── regulatory-impact.sh  # trace regulation to affected matters
+│   │   ├── deadline-audit.sh     # check all deadlines across matters
+│   │   ├── brief-search.sh       # find prior arguments by legal element
+│   │   └── cross-matter.sh       # find patterns across matters
+│   └── tasks/
+│       └── queue.json
+├── self/
+│   ├── identity.md
+│   ├── methodology.md
+│   ├── goals.md
+│   └── diana-profile.md         # practice areas, jurisdictions, preferences
+└── ops/
+    └── methodology/
+        ├── how-precedent-tracking-works.md
+        ├── how-regulatory-propagation-works.md
+        └── how-brief-bank-works.md
+```
+
+---
+
+## Note Schemas
+
+### Case Brief
+
+```yaml
+---
+description: [one sentence — the holding and its significance]
+case_name: "Oracle v. DataShield Technologies"
+citation: "2025 WL 4523891 (9th Cir. 2025)"
+court: 9th-circuit
+court_level: appellate
+jurisdiction: federal
+date_decided: 2025-11-14
+procedural_posture: "appeal from summary judgment"
+issue: "Whether trade secret misappropriation claims based on inevitable disclosure require evidence of actual or threatened misappropriation beyond mere employment by a competitor"
+holding: "Inevitable disclosure alone, without evidence of actual or threatened misappropriation, is insufficient to sustain a trade secret claim under the DTSA"
+reasoning: "The court reasoned that inevitable disclosure doctrine creates an inherent tension with employee mobility rights and cannot substitute for the statute's requirement of actual or threatened misappropriation"
+significance: "Strengthens the 9th Circuit's skepticism toward inevitable disclosure and raises the evidentiary bar for trade secret plaintiffs who cannot prove actual misappropriation"
+dicta: "Court noted in dicta that 'contractual non-compete agreements remain the appropriate vehicle for protecting against competitive employment, not judicial injunctions based on speculative disclosure'"
+treatment:
+  followed_by: []
+  distinguished_by: []
+  overruled_by: null
+  cited_by: ["[[synthex-pi-motion]]"]
+related_cases: ["[[Waymo v. Uber (N.D. Cal. 2018)]]"]
+topics: ["[[trade-secrets]]", "[[precedents]]"]
+relevant_notes: ["[[inevitable-disclosure-doctrine-disfavored-in-ninth-circuit]] — this case strengthens the principle", "[[synthex-trade-secret]] — directly affects our motion strategy"]
+---
+```
+
+### Matter Note
+
+```yaml
+---
+description: [one sentence — matter summary and current posture]
+client: "Synthex Corp"
+matter_name: "Synthex Corp v. NovaTech LLC"
+matter_number: "2026-LIT-0042"
+matter_type: trade-secret-litigation
+status: active
+court: "N.D. Cal."
+jurisdiction: federal
+judge: "Hon. Maria Gonzalez"
+opposing_counsel: "Richards & Morrison LLP (Sarah Park)"
+cause_of_action: ["trade secret misappropriation (DTSA)", "trade secret misappropriation (CUTSA)", "breach of employment agreement"]
+key_facts: "Former Synthex engineer joined NovaTech and allegedly used proprietary ML training pipeline architecture. No non-compete but had confidentiality agreement."
+strategy: "Establish actual misappropriation through document comparison and testimony, avoid reliance on inevitable disclosure given 9th Circuit skepticism"
+key_dates:
+  filed: 2026-01-15
+  answer_due: 2026-02-28
+  initial_case_management: 2026-03-15
+  discovery_cutoff: 2026-09-30
+  dispositive_motions: 2026-11-15
+  trial: 2027-02-01
+deadlines:
+  - date: 2026-02-28
+    description: "Answer and counterclaim deadline"
+    status: pending
+  - date: 2026-03-01
+    description: "PI motion filing deadline"
+    status: drafting
+  - date: 2026-03-15
+    description: "Initial case management conference"
+    status: pending
+related_matters: ["[[techcorp-gdpr-compliance]]"]
+regulations_applicable: ["DTSA (18 U.S.C. 1836)", "CUTSA (Cal. Civ. Code 3426)"]
+work_product: ["[[synthex-pi-motion]]", "[[trade-secret-damages-ninth-circuit]]", "[[synthex-evidence-map]]"]
+topics: ["[[trade-secrets]]", "[[precedents]]"]
+relevant_notes: ["[[inevitable-disclosure-doctrine-disfavored-in-ninth-circuit]] — shapes our strategy: we must prove actual misappropriation", "[[Oracle v. DataShield Technologies]] — key precedent supporting our need for direct evidence"]
+---
+```
+
+### Precedent Principle Note
+
+```yaml
+---
+description: [one sentence — the legal principle and its current status across jurisdictions]
+principle: "inevitable disclosure doctrine"
+jurisdictions:
+  - jurisdiction: 9th-circuit
+    status: disfavored
+    key_case: "[[Oracle v. DataShield Technologies]]"
+    standard: "requires evidence of actual or threatened misappropriation beyond mere competitive employment"
+  - jurisdiction: 7th-circuit
+    status: accepted
+    key_case: "[[PepsiCo v. Redmond (7th Cir. 1995)]]"
+    standard: "inevitable disclosure may support injunctive relief where trade secrets are highly technical and competition is direct"
+  - jurisdiction: california-state
+    status: rejected
+    key_case: "[[Whyte v. Schlage Lock (Cal. Ct. App. 2002)]]"
+    standard: "California strongly disfavors inevitable disclosure as inconsistent with Business and Professions Code 16600"
+last_validated: 2026-02-10
+matters_affected: ["[[synthex-trade-secret]]"]
+topics: ["[[trade-secrets]]", "[[precedents]]"]
+relevant_notes: ["[[Oracle v. DataShield Technologies]] — most recent 9th Circuit treatment", "[[heightened-pleading-applies-to-fraud-based-trade-secret-claims]] — related evidentiary standard in trade secret litigation"]
+---
+```
+
+### Regulatory Update Note
+
+```yaml
+---
+description: [one sentence — what changed and which matters are affected]
+regulation: "GDPR"
+article: "Article 17 — Right to Erasure"
+change_type: amendment | interpretation | enforcement-action | guidance
+effective_date: 2026-02-01
+change_summary: "European Data Protection Board issued new guidance clarifying that legal hold obligations under US litigation do not automatically satisfy the 'legitimate interests' exception to erasure requests"
+source: "EDPB Guidelines 01/2026"
+affected_areas: ["data-retention", "litigation-hold", "cross-border-data"]
+affected_matters: ["[[techcorp-gdpr-compliance]]", "[[meridian-license-dispute]]"]
+required_actions:
+  - matter: "[[techcorp-gdpr-compliance]]"
+    action: "Review litigation hold notice against new EDPB guidance; may need supplemental data protection impact assessment"
+    deadline: 2026-03-01
+    status: pending
+  - matter: "[[meridian-license-dispute]]"
+    action: "Verify data retention clause in license agreement complies with updated guidance"
+    deadline: 2026-03-15
+    status: pending
+propagation_complete: false
+topics: ["[[regulatory]]", "[[privacy-law]]"]
+relevant_notes: ["[[GDPR-article-17-right-to-erasure-overridden-by-legal-hold]] — this principle note needs updating in light of new guidance"]
+---
+```
+
+### Research Memo
+
+```yaml
+---
+description: [one sentence — the legal question addressed and conclusion]
+question_presented: "What is the measure of damages for trade secret misappropriation under DTSA in the Ninth Circuit where the plaintiff seeks lost profits but the trade secret has not been commercialized?"
+short_answer: "The Ninth Circuit permits reasonable royalty as an alternative measure where actual damages cannot be proven, but requires the plaintiff to establish a hypothetical willing-buyer/willing-seller framework grounded in the trade secret's demonstrated or projected market value"
+jurisdictions: ["9th-circuit", "federal"]
+matter: "[[synthex-trade-secret]]"
+authorities_cited:
+  - "[[Oracle v. DataShield Technologies]]"
+  - "[[Waymo v. Uber (N.D. Cal. 2018)]]"
+  - "18 U.S.C. 1836(b)(3)(B)"
+status: final
+drafted: 2026-02-08
+topics: ["[[trade-secrets]]", "[[precedents]]"]
+relevant_notes: ["[[synthex-trade-secret]] — commissioned for this matter", "[[inevitable-disclosure-doctrine-disfavored-in-ninth-circuit]] — shapes the damages theory because we cannot rely on threatened disclosure"]
+---
+```
+
+### Argument Outline
+
+```yaml
+---
+description: [one sentence — the argument and its evidentiary support status]
+matter: "[[synthex-trade-secret]]"
+motion: "preliminary injunction"
+argument: "Defendant's use of identical ML pipeline architecture demonstrates actual misappropriation, not inevitable disclosure"
+legal_standard: "likelihood of success on the merits (Winter v. NRDC)"
+supporting_authorities:
+  - "[[Oracle v. DataShield Technologies]] — establishes need for actual misappropriation evidence"
+  - "18 U.S.C. 1836(b)(3)(A) — injunctive relief provision"
+supporting_evidence:
+  - evidence: "side-by-side code comparison"
+    source: "expert report (Dr. Martinez)"
+    status: obtained
+    strength: strong
+  - evidence: "defendant's access logs during employment"
+    source: "discovery production"
+    status: pending
+    strength: moderate
+  - evidence: "deposition testimony re: architecture knowledge"
+    source: "[[chen-depo-wong-2026-01-28]]"
+    status: obtained
+    strength: strong
+counter_arguments:
+  - argument: "Independent development defense"
+    rebuttal: "Timeline analysis shows NovaTech's pipeline appeared 3 weeks after defendant's start date — insufficient time for independent development of comparable architecture"
+    strength_of_rebuttal: strong
+  - argument: "General industry knowledge defense"
+    rebuttal: "Expert testimony will establish that the specific architecture choices (not the general approach) constitute protectable trade secrets"
+    strength_of_rebuttal: moderate
+gaps: ["need defendant's git commit history to establish timeline"]
+topics: ["[[trade-secrets]]", "[[evidence-standards]]"]
+relevant_notes: ["[[synthex-evidence-map]] — full evidence-to-argument mapping", "[[heightened-pleading-applies-to-fraud-based-trade-secret-claims]] — ensures our pleading standard is met"]
+---
+```
+
+### Deposition Summary
+
+```yaml
+---
+description: [one sentence — the deponent, key testimony, and significance]
+matter: "[[synthex-trade-secret]]"
+deponent: "Dr. James Wong"
+deponent_role: "Former Synthex senior ML engineer, now at NovaTech"
+date: 2026-01-28
+duration_hours: 6.5
+examining_attorney: "Diana Chen"
+key_testimony:
+  - topic: "knowledge of ML pipeline architecture"
+    testimony: "Admitted familiarity with Synthex's proprietary data preprocessing pipeline, including the custom feature engineering module"
+    significance: strong
+    page_line: "pp. 84:12 - 87:3"
+  - topic: "access to source code"
+    testimony: "Confirmed regular access to the full pipeline codebase during employment, including architecture documentation"
+    significance: strong
+    page_line: "pp. 102:15 - 105:8"
+  - topic: "NovaTech development timeline"
+    testimony: "Stated NovaTech's pipeline was 'in development' when he joined, but could not specify what existed before his arrival"
+    significance: moderate
+    page_line: "pp. 145:22 - 148:11"
+admissions: ["knew pipeline architecture was confidential", "did not delete Synthex materials from personal devices until 'sometime after' starting at NovaTech"]
+contradictions: ["claimed no non-compete but employment agreement contains confidentiality clause with non-use provision"]
+impeachment_opportunities: ["testimony about NovaTech timeline contradicts NovaTech's git commit history showing first pipeline commit 18 days after defendant's start date"]
+follow_up_needed: ["obtain defendant's personal device forensics", "subpoena NovaTech git logs for pipeline repository"]
+topics: ["[[trade-secrets]]", "[[evidence-standards]]"]
+relevant_notes: ["[[synthex-evidence-map]] — testimony mapped to legal elements", "[[synthex-pi-motion]] — testimony supports actual misappropriation argument"]
+---
+```
+
+### Compliance Checklist
+
+```yaml
+---
+description: [one sentence — regulation being tracked and current compliance status]
+matter: "[[techcorp-gdpr-compliance]]"
+regulation: "GDPR"
+jurisdiction: EU
+requirements:
+  - requirement: "Data processing inventory (Art. 30)"
+    status: compliant
+    evidence: "Processing register updated 2026-01-15"
+    responsible: "TechCorp DPO"
+    next_review: 2026-04-15
+  - requirement: "Lawful basis documented for each processing activity (Art. 6)"
+    status: compliant
+    evidence: "Lawful basis assessment completed 2025-11-01"
+    responsible: "Diana Chen"
+    next_review: 2026-05-01
+  - requirement: "Data breach notification procedure (Art. 33-34)"
+    status: needs-update
+    evidence: "Procedure drafted but not tested against new EDPB guidance"
+    responsible: "Diana Chen"
+    next_review: 2026-03-01
+  - requirement: "Right to erasure procedure (Art. 17)"
+    status: needs-update
+    evidence: "Current procedure assumes US litigation hold automatically satisfies legitimate interests exception — EDPB 01/2026 guidance changes this"
+    responsible: "Diana Chen"
+    next_review: 2026-03-01
+topics: ["[[regulatory]]", "[[privacy-law]]"]
+relevant_notes: ["[[GDPR-article-17-right-to-erasure-overridden-by-legal-hold]] — principle note being updated", "[[2026-02-01-EDPB-erasure-guidance]] — the regulatory change triggering this update"]
+---
+```
+
+---
+
+## Example Notes
+
+### Case Brief: Oracle v. DataShield Technologies
+
+```markdown
+---
+description: 9th Circuit holds inevitable disclosure alone insufficient for trade secret claim — requires evidence of actual or threatened misappropriation, strengthening employee mobility protections
+case_name: "Oracle v. DataShield Technologies"
+citation: "2025 WL 4523891 (9th Cir. 2025)"
+court: 9th-circuit
+court_level: appellate
+jurisdiction: federal
+date_decided: 2025-11-14
+procedural_posture: "appeal from summary judgment"
+issue: "Whether trade secret misappropriation claims based on inevitable disclosure require evidence of actual or threatened misappropriation beyond mere employment by a competitor"
+holding: "Inevitable disclosure alone, without evidence of actual or threatened misappropriation, is insufficient to sustain a trade secret claim under the DTSA"
+reasoning: "The court reasoned that inevitable disclosure doctrine creates an inherent tension with employee mobility rights and cannot substitute for the statute's requirement of actual or threatened misappropriation"
+significance: "Strengthens the 9th Circuit's skepticism toward inevitable disclosure and raises the evidentiary bar for trade secret plaintiffs who cannot prove actual misappropriation"
+dicta: "Court noted in dicta that contractual non-compete agreements remain the appropriate vehicle for protecting against competitive employment, not judicial injunctions based on speculative disclosure"
+treatment:
+  followed_by: []
+  distinguished_by: []
+  overruled_by: null
+  cited_by: ["[[synthex-pi-motion]]"]
+related_cases: ["[[Waymo v. Uber (N.D. Cal. 2018)]]", "[[PepsiCo v. Redmond (7th Cir. 1995)]]"]
+topics: ["[[trade-secrets]]", "[[precedents]]"]
+relevant_notes: ["[[inevitable-disclosure-doctrine-disfavored-in-ninth-circuit]] — this case provides the strongest 9th Circuit statement against inevitable disclosure", "[[synthex-trade-secret]] — directly shapes our motion strategy: we must establish actual misappropriation through document comparison, not speculative disclosure"]
+---
+
+# Oracle v. DataShield Technologies
+
+## Facts
+
+Oracle employed a senior database engineer for eight years. The engineer left to join DataShield Technologies, a direct competitor developing a competing database optimization product. Oracle sought a preliminary injunction under the DTSA, arguing that the engineer's knowledge of Oracle's proprietary optimization algorithms made disclosure inevitable in her new role. Oracle did not allege that the engineer had taken any documents, code, or proprietary materials.
+
+## Analysis
+
+The court's reasoning turns on the statutory text of the DTSA, which requires "actual or threatened misappropriation" as a predicate for relief. The court found that inevitable disclosure theory stretches "threatened misappropriation" beyond its statutory meaning by converting specialized knowledge — which employees legitimately develop through their work — into a basis for restricting future employment. The court distinguished [[PepsiCo v. Redmond (7th Cir. 1995)]], noting that the 7th Circuit's adoption of inevitable disclosure predated the DTSA and relied on state trade secret law with different statutory language.
+
+The practical consequence is significant for trade secret plaintiffs in the 9th Circuit: without evidence of actual misappropriation (document theft, code copying, data exfiltration) or specific, articulable threats of disclosure (not mere competitive employment), injunctive relief under the DTSA is unavailable. This raises the evidentiary burden and pushes plaintiffs toward contractual remedies (non-compete agreements, which are themselves disfavored in California under Bus. & Prof. Code 16600).
+
+## Impact on Active Matters
+
+This case directly affects [[synthex-trade-secret]]. Our motion strategy must establish actual misappropriation through the code comparison evidence and deposition testimony in [[chen-depo-wong-2026-01-28]], not through an inevitable disclosure theory. The fact that we have document comparison evidence (identical ML pipeline architecture) and timeline evidence (pipeline appeared 18 days after defendant's start) puts us in a stronger position than Oracle was — but we must present this as actual misappropriation, not inevitable disclosure.
+
+Since [[heightened-pleading-applies-to-fraud-based-trade-secret-claims]], our pleading must satisfy the higher specificity standard. The code comparison evidence should be described with architectural detail, not conclusory allegations.
+
+---
+```
+
+### Precedent Principle: Inevitable Disclosure in the 9th Circuit
+
+```markdown
+---
+description: The 9th Circuit disfavors inevitable disclosure as a standalone basis for trade secret claims — three key cases establish the requirement for evidence of actual or threatened misappropriation
+principle: "inevitable disclosure doctrine"
+jurisdictions:
+  - jurisdiction: 9th-circuit
+    status: disfavored
+    key_case: "[[Oracle v. DataShield Technologies]]"
+    standard: "requires evidence of actual or threatened misappropriation beyond mere competitive employment"
+  - jurisdiction: 7th-circuit
+    status: accepted
+    key_case: "[[PepsiCo v. Redmond (7th Cir. 1995)]]"
+    standard: "inevitable disclosure may support injunctive relief where trade secrets are highly technical and competition is direct"
+  - jurisdiction: california-state
+    status: rejected
+    key_case: "[[Whyte v. Schlage Lock (Cal. Ct. App. 2002)]]"
+    standard: "California strongly disfavors inevitable disclosure as inconsistent with Business and Professions Code 16600"
+last_validated: 2026-02-10
+matters_affected: ["[[synthex-trade-secret]]"]
+topics: ["[[trade-secrets]]", "[[precedents]]"]
+relevant_notes: ["[[Oracle v. DataShield Technologies]] — the strongest 9th Circuit statement against inevitable disclosure", "[[heightened-pleading-applies-to-fraud-based-trade-secret-claims]] — related evidentiary standard that compounds the burden on trade secret plaintiffs", "[[contractual-limitation-of-liability-unenforceable-for-willful-breach]] — the contractual alternative when injunctive relief is unavailable"]
+---
+
+# inevitable-disclosure-doctrine-disfavored-in-ninth-circuit
+
+## The Principle
+
+The inevitable disclosure doctrine posits that a court may enjoin a former employee from working for a competitor when the employee's knowledge of trade secrets makes disclosure inevitable in the new role, even absent evidence that the employee has actually taken or used proprietary information. The doctrine exists in tension with employee mobility rights and varies dramatically across jurisdictions.
+
+## Jurisdictional Treatment
+
+In the 9th Circuit, since [[Oracle v. DataShield Technologies]], inevitable disclosure standing alone is insufficient to sustain a trade secret misappropriation claim under the DTSA. The court requires evidence of actual or threatened misappropriation — defined as specific, articulable conduct beyond mere competitive employment. This aligns with California's strong public policy favoring employee mobility and its statutory prohibition against non-compete agreements (Bus. & Prof. Code 16600).
+
+The 7th Circuit takes the opposite approach. Since [[PepsiCo v. Redmond (7th Cir. 1995)]], inevitable disclosure can support injunctive relief where the trade secrets are highly technical, the competition is direct, and the former employer can show that the employee would be unable to perform the new job without drawing on protected knowledge. Several district courts in the 7th Circuit have applied this standard, though the doctrine has been narrowed in subsequent decisions.
+
+California state courts reject inevitable disclosure entirely. Since [[Whyte v. Schlage Lock (Cal. Ct. App. 2002)]], the doctrine is viewed as an end-run around California's prohibition on non-compete agreements. A trade secret plaintiff in California state court must prove actual misappropriation or a concrete, imminent threat of disclosure — hypothetical or speculative disclosure based on job similarity is not enough.
+
+## Practical Implications
+
+For matters in the 9th Circuit or California state courts, the strategy must be built on actual misappropriation evidence: documents taken, code copied, data exfiltrated, or testimony admitting use. The [[synthex-trade-secret]] matter has strong actual misappropriation evidence (code comparison showing identical architecture, timeline showing impossibility of independent development), which is exactly the type of case that survives the 9th Circuit's skepticism.
+
+For matters in the 7th Circuit or jurisdictions that accept inevitable disclosure, the doctrine remains a viable supplemental theory, but should not be the primary basis for relief. Even in accepting jurisdictions, courts have narrowed the doctrine and imposed additional requirements.
+
+## Monitoring
+
+The agent validates this principle quarterly by checking Westlaw/Lexis for new 9th Circuit decisions citing [[Oracle v. DataShield Technologies]] or applying inevitable disclosure analysis. Any case that follows, distinguishes, or overrules Oracle gets briefed and linked here. As of the last validation (Feb 10, 2026), Oracle has been cited favorably in two district court decisions and has not been distinguished or overruled.
+
+---
+```
+
+### Regulatory Update: EDPB Erasure Guidance
+
+```markdown
+---
+description: New EDPB guidance narrows the legal hold exception to GDPR right-to-erasure requests — affects two active matters requiring immediate compliance review
+regulation: "GDPR"
+article: "Article 17 — Right to Erasure"
+change_type: guidance
+effective_date: 2026-02-01
+change_summary: "European Data Protection Board issued new guidance clarifying that legal hold obligations under US litigation do not automatically satisfy the legitimate interests exception to erasure requests — requires case-by-case balancing test with documented justification"
+source: "EDPB Guidelines 01/2026"
+affected_areas: ["data-retention", "litigation-hold", "cross-border-data"]
+affected_matters: ["[[techcorp-gdpr-compliance]]", "[[meridian-license-dispute]]"]
+required_actions:
+  - matter: "[[techcorp-gdpr-compliance]]"
+    action: "Review litigation hold notice against new EDPB guidance; conduct supplemental data protection impact assessment for US-held data subject to both litigation hold and GDPR erasure requests"
+    deadline: 2026-03-01
+    status: pending
+  - matter: "[[meridian-license-dispute]]"
+    action: "Verify that data retention clause in license agreement does not conflict with updated erasure guidance; review DPA addendum"
+    deadline: 2026-03-15
+    status: pending
+propagation_complete: false
+topics: ["[[regulatory]]", "[[privacy-law]]"]
+relevant_notes: ["[[GDPR-article-17-right-to-erasure-overridden-by-legal-hold]] — this principle note must be updated to reflect the new guidance", "[[techcorp-gdpr-compliance]] — most urgent affected matter: client has active litigation hold AND GDPR exposure"]
+---
+
+# 2026-02-01 EDPB Erasure Guidance Update
+
+## What Changed
+
+The EDPB issued Guidelines 01/2026 on February 1, clarifying the interaction between the GDPR right to erasure (Article 17) and litigation hold obligations under non-EU law. The key change: US litigation holds do not automatically satisfy the "legitimate interests" exception to erasure requests under Article 6(1)(f). Instead, controllers must conduct a case-by-case balancing test weighing:
+
+1. The necessity of data retention for litigation purposes
+2. The data subject's reasonable expectations regarding erasure
+3. The specificity of the litigation hold (broad holds covering entire databases receive less deference than targeted holds on identified documents)
+4. Whether the data can be pseudonymized during the hold period
+
+## Why This Matters
+
+Previously, our standard advice to clients was that a valid US litigation hold provided sufficient legal basis to deny GDPR erasure requests for held data. Since [[GDPR-article-17-right-to-erasure-overridden-by-legal-hold]], this was the established principle. The new guidance does not eliminate this position but requires additional documentation: a written balancing test for each category of held data, regular reassessment of hold scope, and pseudonymization where feasible.
+
+## Impact on Active Matters
+
+**[[techcorp-gdpr-compliance]] — HIGH PRIORITY:** TechCorp is currently under US litigation hold (Chen v. DataFlow matter) while also processing EU data subjects' personal data. The existing litigation hold notice does not include the EDPB-required balancing test. Diana needs to:
+1. Review the hold scope for necessity and specificity
+2. Draft a supplemental DPIA for data subject to both obligations
+3. Implement pseudonymization where hold does not require identifying data
+4. Document the balancing test and make it available for supervisory authority review
+
+**[[meridian-license-dispute]] — MODERATE PRIORITY:** The Meridian license agreement contains a data retention clause that may conflict with updated erasure guidance. The DPA addendum references "legitimate interests" as the lawful basis for post-termination data retention. Under the new guidance, this reference alone is insufficient — the DPA addendum needs a documented balancing test.
+
+## Principle Note Update Required
+
+The note [[GDPR-article-17-right-to-erasure-overridden-by-legal-hold]] must be updated to reflect the narrowing of this principle. The title may need revision: "overridden by" is no longer accurate — "qualified by" or "subject to balancing test alongside" better captures the current state.
+
+---
+```
+
+### Matter Note: Synthex Trade Secret
+
+```markdown
+---
+description: Active trade secret litigation in N.D. Cal — PI motion deadline March 1, strategy centers on actual misappropriation evidence to avoid inevitable disclosure reliance
+client: "Synthex Corp"
+matter_name: "Synthex Corp v. NovaTech LLC"
+matter_number: "2026-LIT-0042"
+matter_type: trade-secret-litigation
+status: active
+court: "N.D. Cal."
+jurisdiction: federal
+judge: "Hon. Maria Gonzalez"
+opposing_counsel: "Richards & Morrison LLP (Sarah Park)"
+cause_of_action: ["trade secret misappropriation (DTSA)", "trade secret misappropriation (CUTSA)", "breach of employment agreement"]
+key_facts: "Former Synthex engineer (Dr. James Wong) joined NovaTech and allegedly used proprietary ML training pipeline architecture. No non-compete but had confidentiality agreement with non-use provision."
+strategy: "Establish actual misappropriation through side-by-side code comparison (expert report) and timeline evidence (pipeline appeared 18 days after defendant's start date). Avoid reliance on inevitable disclosure given 9th Circuit skepticism per Oracle v. DataShield."
+key_dates:
+  filed: 2026-01-15
+  answer_due: 2026-02-28
+  pi_motion_deadline: 2026-03-01
+  initial_case_management: 2026-03-15
+  discovery_cutoff: 2026-09-30
+  dispositive_motions: 2026-11-15
+  trial: 2027-02-01
+deadlines:
+  - date: 2026-02-28
+    description: "Answer and counterclaim deadline"
+    status: pending
+  - date: 2026-03-01
+    description: "PI motion filing deadline"
+    status: drafting
+  - date: 2026-03-15
+    description: "Initial case management conference"
+    status: pending
+related_matters: []
+regulations_applicable: ["DTSA (18 U.S.C. 1836)", "CUTSA (Cal. Civ. Code 3426)"]
+work_product: ["[[synthex-pi-motion]]", "[[trade-secret-damages-ninth-circuit]]", "[[synthex-evidence-map]]", "[[chen-depo-wong-2026-01-28]]"]
+topics: ["[[trade-secrets]]", "[[precedents]]"]
+relevant_notes: ["[[inevitable-disclosure-doctrine-disfavored-in-ninth-circuit]] — why our strategy avoids this theory", "[[Oracle v. DataShield Technologies]] — controlling precedent on the evidentiary standard", "[[heightened-pleading-applies-to-fraud-based-trade-secret-claims]] — affects our pleading specificity requirements"]
+---
+
+# Synthex Corp v. NovaTech LLC
+
+## Matter Summary
+
+Synthex alleges that Dr. James Wong, a former senior ML engineer, misappropriated Synthex's proprietary machine learning training pipeline architecture after leaving to join competitor NovaTech. The evidence suggests actual misappropriation rather than theoretical inevitable disclosure: NovaTech's pipeline appeared in their codebase 18 days after Wong's start date, the architecture is structurally identical to Synthex's proprietary system, and Wong admitted in deposition (see [[chen-depo-wong-2026-01-28]]) to having had access to the full pipeline codebase and architecture documentation during his employment.
+
+## Strategy
+
+Since [[inevitable-disclosure-doctrine-disfavored-in-ninth-circuit]], our strategy must rely on actual misappropriation evidence, not speculative disclosure. This works in our favor because we have strong actual misappropriation evidence:
+
+1. **Code comparison:** Expert report (Dr. Martinez) shows architecturally identical pipeline with specific implementation choices that are not standard in the industry
+2. **Timeline:** NovaTech's first git commit for the pipeline is 18 days after Wong's start date — insufficient time for independent development of comparable architecture
+3. **Deposition testimony:** Wong admitted familiarity with and access to the proprietary pipeline, did not delete Synthex materials from personal devices until "sometime after" starting at NovaTech (see [[chen-depo-wong-2026-01-28]])
+4. **Confidentiality agreement:** Wong signed an agreement with a non-use provision — breach of this agreement is an independent cause of action
+
+Since [[Oracle v. DataShield Technologies]], the 9th Circuit requires specific, articulable evidence of actual or threatened misappropriation. Our evidence meets this standard. The PI motion ([[synthex-pi-motion]]) is being drafted with this framework.
+
+## Evidence Gaps
+
+- Need defendant's personal device forensics (motion to compel may be required)
+- NovaTech git commit logs for pipeline repository (subpoena pending)
+- Expert testimony on the non-standard nature of the specific architecture choices (Dr. Martinez report in progress)
+
+## Key Risk
+
+NovaTech's strongest defense is independent development. If they can show that any portion of the pipeline existed before Wong's arrival, our timeline argument weakens. The git commit history is critical — if the first commit predates Wong's start date, we lose the timeline argument entirely.
+
+---
+```
+
+### Evidence Map: Synthex
+
+```markdown
+---
+description: Maps every piece of evidence in Synthex v. NovaTech to the legal elements it supports — identifies evidentiary gaps and strength of proof per element
+matter: "[[synthex-trade-secret]]"
+claim: "trade secret misappropriation (DTSA)"
+last_updated: 2026-02-14
+overall_strength: strong
+topics: ["[[trade-secrets]]", "[[evidence-standards]]"]
+relevant_notes: ["[[synthex-pi-motion]] — this map informs the argument structure", "[[chen-depo-wong-2026-01-28]] — primary testimonial evidence"]
+---
+
+# Synthex Evidence Map
+
+## Element 1: Existence of a Trade Secret
+
+| Evidence | Source | Status | Strength |
+|----------|--------|--------|----------|
+| ML pipeline architecture documentation | Synthex production | obtained | strong |
+| Expert report: architecture is not standard practice | Dr. Martinez | in progress | strong (expected) |
+| Confidentiality agreement designating pipeline as trade secret | Synthex HR records | obtained | strong |
+| Wong testimony: knew pipeline was confidential | [[chen-depo-wong-2026-01-28]] p.84 | obtained | strong |
+
+**Assessment:** Strong on this element. Multiple independent sources confirm the pipeline architecture was treated as confidential and has independent economic value from not being generally known.
+
+## Element 2: Misappropriation (Acquisition or Use)
+
+| Evidence | Source | Status | Strength |
+|----------|--------|--------|----------|
+| Side-by-side code comparison showing identical architecture | Expert report (Martinez) | in progress | strong (expected) |
+| NovaTech pipeline first commit 18 days after Wong's start | NovaTech git logs | subpoena pending | strong (if obtained) |
+| Wong admitted access to full pipeline codebase | [[chen-depo-wong-2026-01-28]] p.102 | obtained | strong |
+| Wong did not delete Synthex materials until "sometime after" start | [[chen-depo-wong-2026-01-28]] p.156 | obtained | moderate |
+| Personal device forensics showing Synthex files | Motion to compel | not yet filed | strong (if obtained) |
+
+**Assessment:** Currently moderate-to-strong. The code comparison expert report (expected late February) will be the linchpin. Git commit timeline is critical for the independent development rebuttal. Without either, this element weakens to the testimony alone, which is circumstantial.
+
+**Gap:** Need the git commit history. If NovaTech destroys or claims privilege over repository logs, we need a backup approach — potentially through Wong's personal commit history or NovaTech's CI/CD deployment records.
+
+## Element 3: Damages or Threat of Irreparable Harm (for PI)
+
+| Evidence | Source | Status | Strength |
+|----------|--------|--------|----------|
+| NovaTech marketing materials advertising pipeline capabilities | Public filings | obtained | moderate |
+| Customer declarations re: competitive impact | Synthex clients | 2 obtained, 1 pending | moderate |
+| Expert report on reasonable royalty valuation | Economic expert | not yet retained | pending |
+
+**Assessment:** Weakest element. For the PI motion, irreparable harm is the standard — but since [[Oracle v. DataShield Technologies]], mere competitive harm without specific evidence of ongoing misappropriation may be insufficient. We need to show that NovaTech is actively using the misappropriated architecture in its current product, not just that they acquired it.
+
+---
+```
+
+---
+
+## Processing Workflow
+
+### Capture
+
+Legal knowledge enters the system through multiple channels:
+
+1. **New case law:** Cases identified through legal research alerts, opposing counsel citations, or CLE materials land in `00_inbox/cases/` for briefing.
+2. **Regulatory changes:** Regulatory updates, agency guidance, and enforcement actions land in `00_inbox/regulations/` for impact assessment.
+3. **Matter events:** Deposition transcripts, filed motions, discovery productions, court orders — each generates a note linked to the matter file.
+4. **Research:** Legal articles, treatises, and CLE materials land in `00_inbox/research/` for claim extraction.
+5. **Contracts:** Agreements for review land in `00_inbox/contracts/`.
+
+### Process (the domain-specific step)
+
+Four processing workflows, each with distinct legal requirements:
+
+**Case briefing pipeline:**
+1. Case arrives in inbox → agent creates case brief note from template
+2. Agent extracts: issue, holding, reasoning, dicta, significance
+3. Agent checks if case affects existing precedent principles → updates treatment chains (followed_by, distinguished_by, overruled_by)
+4. Agent checks if case affects active matters → links and alerts
+5. Agent creates or updates precedent principle notes
+
+**Regulatory impact pipeline:**
+1. Regulatory change identified → agent creates regulatory update note
+2. Agent traces affected areas through YAML metadata across all matter notes and compliance checklists
+3. Agent generates required_actions for each affected matter with deadlines
+4. Agent flags affected precedent principle notes for review
+5. Agent sets propagation_complete to false until all affected matters have been updated
+
+**Argument construction pipeline:**
+1. Legal question identified → agent searches brief bank for prior work product addressing similar issues
+2. Agent searches precedent graph for supporting authorities
+3. Agent creates argument outline with evidence mapping
+4. Agent identifies evidentiary gaps — what evidence is needed but not yet obtained
+5. Agent cross-references counter-arguments from opposing counsel's prior briefs (if in the system)
+
+**Matter review pipeline (weekly):**
+1. Agent reviews all active matter deadlines → flags approaching deadlines
+2. Agent checks all cited precedents against treatment chains → alerts if any relied-upon case has been subsequently distinguished or overruled
+3. Agent checks all applicable regulations for recent changes → cross-references against regulatory update notes
+4. Agent reviews evidence maps for gaps → prioritizes discovery and investigation tasks
+5. Agent generates weekly matter status summary
+
+### Connect
+
+Every case brief links to the principles it establishes or modifies. Every principle links to every matter it affects. Every regulatory update links to every matter and principle it impacts. Every argument links to the evidence that supports it. The agent maintains these connections bidirectionally and continuously — when a new case is briefed, it propagates through the precedent chain, through affected matters, and through any arguments that relied on the modified principle. This is the graph advantage: a change anywhere ripples everywhere it matters.
+
+### Verify
+
+Weekly: precedent chain validation (are cited cases still good law?), deadline audit (are any deadlines approaching without assigned tasks?), evidence map gap check.
+Monthly: compliance checklist review, regulatory landscape scan, cross-matter pattern analysis.
+Quarterly: full precedent graph validation, brief bank indexing, matter archival for closed matters.
+
+---
+
+## MOC Structure
+
+### Hub MOC: index.md
+
+```markdown
+---
+description: Entry point for Diana's legal knowledge system — navigate to precedent law, regulatory landscape, active matters, or legal work product
+type: moc
+---
+
+# Legal Practice
+
+## Active Navigation
+- [[precedents]] — case law principles, treatment chains, jurisdictional variations
+- [[regulatory]] — regulatory landscape, compliance tracking, change propagation
+- [[trade-secrets]] — trade secret law, inevitable disclosure, misappropriation standards
+- [[privacy-law]] — privacy and data protection, GDPR, CCPA, cross-border data
+
+## Practice Areas
+- [[contracts]] — contract law patterns, interpretation, limitation of liability
+- [[ip-licensing]] — IP licensing, fair use, transformative use
+- [[evidence-standards]] — evidentiary standards, pleading requirements, burden of proof
+- [[procedural]] — procedural rules, deadlines, jurisdictional requirements
+
+---
+
+Agent Notes:
+When a new case is briefed, always check [[precedents]] for affected principle notes before filing. When a regulatory change arrives, always run the impact trace through all active matters before marking it processed. The precedent chain and regulatory propagation are the two highest-value automated workflows in this system.
+```
+
+### Domain MOC: precedents.md
+
+```markdown
+---
+description: Case law principles with treatment tracking — where every relied-upon holding is validated against its citation chain and jurisdictional variations are mapped
+type: moc
+topics: ["[[index]]"]
+---
+
+# precedents
+
+This MOC tracks legal principles established by case law, organized by practice area and jurisdiction. The critical function is treatment tracking: when a case is overruled, distinguished, or limited, every downstream note that relies on that case gets flagged.
+
+## Trade Secret Principles
+- [[inevitable-disclosure-doctrine-disfavored-in-ninth-circuit]] — the doctrine is disfavored in the 9th Circuit and rejected in California; 7th Circuit accepts with limitations
+- [[heightened-pleading-applies-to-fraud-based-trade-secret-claims]] — fraud-based trade secret claims must satisfy Rule 9(b) specificity requirements
+
+## Privacy Principles
+- [[CCPA-private-right-of-action-limited-to-data-breaches]] — CCPA private right only covers data breach claims, not all privacy violations
+- [[GDPR-article-17-right-to-erasure-overridden-by-legal-hold]] — NEEDS UPDATE: EDPB 01/2026 guidance narrows this principle
+
+## Contract Principles
+- [[contractual-limitation-of-liability-unenforceable-for-willful-breach]] — limitation of liability clauses generally unenforceable where breach is willful or grossly negligent
+- [[transformative-use-test-requires-new-expression-not-new-purpose]] — transformative use in copyright fair use analysis
+
+## Treatment Alerts
+- **ACTIVE:** [[GDPR-article-17-right-to-erasure-overridden-by-legal-hold]] — principle narrowed by EDPB 01/2026 guidance. Affects [[techcorp-gdpr-compliance]] and [[meridian-license-dispute]]. Update pending.
+
+## Validation Schedule
+Last full validation: 2026-02-10. Next scheduled: 2026-05-10. The agent checks all cited cases quarterly against Westlaw/Lexis alerts for subsequent treatment.
+
+---
+
+Agent Notes:
+Treatment alerts are the highest-priority items in this MOC. Any principle in the Treatment Alerts section requires immediate attention — it means a relied-upon case has been affected by subsequent authority. Check affected matters immediately and assess whether pending or filed briefs need supplemental authority notices.
+```
+
+---
+
+## Graph Query Examples
+
+```bash
+# Find all matters affected by a specific regulation
+rg '^regulations_applicable:.*GDPR' vault/03_matters/
+
+# Find all case briefs from a specific court
+rg '^court: 9th-circuit' vault/01_thinking/ vault/04_legal/
+
+# Find all arguments with evidentiary gaps
+rg '^gaps:' vault/04_legal/briefs/ | grep -v '\[\]'
+
+# Find all matters with approaching deadlines
+rg '^\s*- date: 2026-02' vault/03_matters/ -A 2
+
+# Find all precedent principles that need updating
+rg '^last_validated:' vault/01_thinking/ | \
+  while read line; do
+    date=$(echo "$line" | grep -o '20[0-9][0-9]-[0-9][0-9]-[0-9][0-9]')
+    if [[ "$date" < "2025-11-15" ]]; then echo "STALE: $line"; fi
+  done
+
+# Cross-matter pattern: find all matters using the same cause of action
+rg '^cause_of_action:.*trade.secret' vault/03_matters/
+
+# Find all deposition contradictions (impeachment opportunities)
+rg '^impeachment_opportunities:' vault/04_legal/depositions/
+
+# Track treatment of a specific case
+rg 'Oracle v. DataShield' vault/ --glob '*.md'
+
+# Find all regulatory changes with incomplete propagation
+rg '^propagation_complete: false' vault/01_thinking/ vault/04_legal/
+
+# Brief bank search: find prior work on a legal issue
+rg '^question_presented:.*trade.secret.*damages' vault/04_legal/research-memos/
+```
+
+---
+
+## What Makes This Domain Unique
+
+### Precedent chain tracking is a graph integrity problem
+
+Legal reasoning depends on the continued validity of prior authority. A case cited in a brief is only as good as its current treatment: if it has been overruled, distinguished, or limited by a subsequent decision, the argument built on it is weakened or destroyed. This is not a "nice to have" — citing overruled authority is a sanctionable offense in federal courts.
+
+The knowledge graph turns this from a manual research task into a structural property. Every case brief has a `treatment` field tracking how it has been followed, distinguished, or overruled. Every argument outline lists its supporting authorities. When a new case is briefed and the agent detects that it distinguishes or overrules an existing case in the graph, it traces forward through every argument and matter that relied on the affected case and generates alerts. No attorney can maintain this level of citation hygiene manually across 15 active matters spanning multiple practice areas. The graph makes it structural.
+
+### Regulatory change propagation across matters is the cross-cutting killer feature
+
+When the EDPB issues new guidance on data erasure, Diana needs to know which of her 15 matters are affected. She might remember the obvious one (the GDPR compliance matter) but miss the less obvious one (the license dispute where the DPA addendum references "legitimate interests"). The agent traces regulatory changes through every matter's `regulations_applicable` field and every compliance checklist's requirements, generating action items for each affected engagement. This propagation is instantaneous and exhaustive — it catches the connections that human memory misses.
+
+### Dense schema enables graph database queries over documents
+
+Legal work product is inherently structured: citations have formats, courts have hierarchies, jurisdictions have boundaries, causes of action have elements, evidence has admissibility standards. The dense YAML schema captures this structure so that the vault is queryable like a database. "Find all 9th Circuit decisions on trade secret misappropriation where the court applied a heightened pleading standard" is a ripgrep query, not a research project. This transforms the brief bank from a keyword-searchable file dump into a structured legal knowledge base where every dimension of every artifact is queryable independently.
+
+---
+
+## Agent-Native Advantages
+
+### Exhaustive precedent chain validation across all active matters
+
+The agent validates every cited case in every active matter against its current treatment status on a quarterly cycle. For each case in the precedent graph, it checks whether subsequent decisions have followed, distinguished, limited, or overruled the holding. If a relied-upon case has been affected, the agent traces forward through every argument outline, research memo, and matter note that cites it, generating specific alerts: "Case X, relied upon in your PI motion for matter Y, was distinguished by case Z last month on the scope-of-disclosure issue. Review whether the distinguishing factors apply to your facts."
+
+**What this looks like in practice:** Diana's precedent graph contains 85 case briefs across four practice areas. A quarterly validation run checks each case's treatment status. The agent detects that a case she relied upon in the Synthex PI motion was cited (but not distinguished) in a new district court decision. It reads the new decision, determines that the citation is favorable (following, not distinguishing), and adds the citation to the case brief's `treatment.cited_by` field. But it also detects that a case cited in the Meridian contract dispute was limited by a new 2nd Circuit decision. It alerts Diana: "The 2nd Circuit limited the holding in [case] to cases involving software licensing specifically. Your Meridian matter involves hardware licensing. Review whether this limitation affects your argument at paragraph 23 of the draft motion." Diana's DMS could never do this because it treats documents as text blobs, not as nodes in a precedent chain graph.
+
+### Cross-matter pattern detection and work product reuse
+
+The agent maintains awareness of all arguments, research memos, and legal strategies across all matters — active and archived. When Diana begins research for a new matter, the agent searches the brief bank not by keyword but by legal element: "Has Diana previously argued trade secret misappropriation damages in the 9th Circuit? What authorities did she cite? What counter-arguments did she face?" The search is semantic and structural, leveraging the YAML metadata to find relevant work product that keyword search would miss.
+
+**What this looks like in practice:** Diana opens a new matter involving breach of a software licensing agreement. The agent searches the brief bank and finds: "You drafted a research memo on contractual limitation of liability in the Meridian matter (2025) that analyzed 9th Circuit standards. The Meridian memo cited 4 cases that directly apply to the new matter's liability cap issue. You also drafted an argument outline for a similar clause in the ACME matter (2024) that addressed the willful breach exception. Both are relevant to the new matter. Shall I create an argument outline for the new matter pre-populated with these authorities and adapted for the current facts?" Diana would have spent 3 hours re-researching what she already knew. The agent found it in seconds because the YAML structure (jurisdiction, legal issue, authorities cited) made the connection visible.
+
+### Regulatory change impact assessment with matter-level granularity
+
+When a regulatory change occurs, the agent performs instant impact assessment across all matters, compliance checklists, and contractual obligations. The propagation is not a keyword search for the regulation's name — it traces through the regulation's `affected_areas` to every matter whose `regulations_applicable` includes the changed regulation, every compliance checklist whose requirements reference the changed provision, and every contract review note whose terms interact with the changed rule. The agent generates specific, actionable items for each affected engagement with proposed deadlines.
+
+**What this looks like in practice:** The EDPB issues new erasure guidance on February 1. Within minutes of the regulatory update note being created, the agent has identified every affected matter: TechCorp (active litigation hold + GDPR exposure), Meridian (DPA addendum with "legitimate interests" reference), and a third matter Diana did not immediately consider (a data processing agreement review in the inbox that references the same Article 17 exception). The agent generates three action items with tiered deadlines based on urgency, flags the relevant precedent principle note for update, and marks propagation as incomplete until Diana confirms each matter has been addressed. Diana's previous system — a regulatory alert email that she flagged and meant to trace through her matters this weekend — would have caught TechCorp because it was obvious, probably caught Meridian because she remembered, and missed the inbox item entirely. The agent catches all three because exhaustive propagation is what agents do.
+
+### Deadline monitoring with jurisdictional calculation
+
+Legal deadlines are not simple calendar entries. They interact with procedural rules, court schedules, holiday calendars, and jurisdictional variations. Filing deadlines in federal court are calculated differently than state court. Service rules vary by jurisdiction. Tolling provisions affect statutes of limitations. The agent calculates and monitors all deadlines across all matters simultaneously, accounting for these jurisdictional variations.
+
+**What this looks like in practice:** The agent maintains a rolling deadline report across all 15 active matters. It flags: "Three deadlines within the next 14 days. (1) Synthex PI motion filing — March 1, 14 days remaining, status: drafting, evidence gap: expert report pending. (2) TechCorp GDPR compliance review — March 1, 14 days remaining, triggered by EDPB guidance. (3) Meridian answer — March 5, 18 days remaining, status: not started." Each deadline includes its calculation basis (procedural rule + filing), dependent tasks (what must be completed before the deadline can be met), and risk assessment (which deadlines are at risk based on the status of dependent tasks). No calendar application can provide this level of procedural awareness because calendars track dates, not the legal rules and dependent tasks that determine whether those dates can be met.
+
+---
+---
+
+Topics:
+- [[domain-compositions]]