Sets up the Socrata project repo with: docs/ — strategy and design documents - idea.md: full product vision - implementation-plan.md: Phase 0 + Phase 1 MVP plan - phase-0-validation.md: 2-week validation experiment strategy - phase-0-plan.md: concrete Phase 0 build plan - phase-0-results.md: Phase 0 gate outcome — GO for MVP - sysml-modeling.md: metamodel + SE discipline + validation rules - socrates.md: agent character, surfaces, modes, prompts, lifecycle - sync.md: bidirectional text↔diagram sync engineering - design-source/: HTML/CSS/JS handoff bundle from Claude Design phase-0/ — validated harness (CLI, no UI, no DB) - LM Studio (local OpenAI-compatible) generation + detection + judge - PlantUML rendering for SysML model visualization - 10-seed corpus (8 working + 2 holdouts) - 5 corpus runs with iteration history in reports/ - Final gate: 10/10 pass, mean 4.32/5, holdouts validated Phase 1 MVP scope and milestones documented in implementation-plan.md.
20 lines
1.3 KiB
JSON
20 lines
1.3 KiB
JSON
{
|
||
"id": "03-redline-ai",
|
||
"title": "Redline AI",
|
||
"problem": "In-house legal teams spend hours per contract on first-pass redlining — tedious mechanical work that doesn't require senior judgment but does require consistent application of company-specific clause libraries. Junior associates burn out doing it; outside counsel charges $400/hr for it.",
|
||
"targetUser": "General Counsel and Legal Operations leads at companies with 50–500 employees that handle 100+ contracts per quarter (SaaS, marketplaces, services).",
|
||
"desiredOutcome": "First-pass redlines on standard contracts complete in minutes, with the GC reviewing AI suggestions instead of authoring redlines from scratch.",
|
||
"initialHypothesis": "Legal teams will trust AI for first-pass mechanical redlining but reserve final judgment for humans — i.e., the bar is 'better than a tired junior associate', not 'better than the GC'.",
|
||
"constraints": [
|
||
"Must integrate with Microsoft Word — that's where the work happens",
|
||
"Contract content must never train cross-customer models (privilege concerns)",
|
||
"Audit trail of every change suggestion must be exportable",
|
||
"SOC 2 Type II compliance from day one"
|
||
],
|
||
"meta": {
|
||
"isHoldout": false,
|
||
"expectedDifficulty": "medium",
|
||
"testsFor": ["b2b", "domain-heavy", "regulatory", "integration"]
|
||
}
|
||
}
|