Stonewall Brain v10.4
SKILL.md defines orchestration, retrieval priorities, and operational guardrails.
Six companion skills handle alt-voice, maintenance, depo prep, transcription, and celebration.
What Stonewall does, how it works, and what it looks like when the system is running at operational tempo.
The Stonewall system is a multi-surface corpus engine. Raw evidence flows in from OneDrive and email exports, gets normalized into searchable markdown, and syncs to Notion for case-level relations and workflow fields. The Brain skill handles recall, the codices enforce canon, and the catalog tracks every artifact from intake to analysis.
SKILL.md defines orchestration, retrieval priorities, and operational guardrails.
Six companion skills handle alt-voice, maintenance, depo prep, transcription, and celebration.
catalog/manifest.md is the primary artifact index — searchable by character,
pattern, date, case, or keyword. Five derived index files provide alternative access paths.
Cast codex, case codex, artifact codex, phenomenology registry, and deposition codex govern identity, chronology, patterns, and matter posture.
PDF sidecar generation, OneDrive intake, email consolidation, Notion sync, manifest backfill, and the Hoss Shalom capture pipeline — all stdlib Python and Node.
The control plane spans the repo, Notion, and external evidence reservoirs. These are the living surfaces, not historical snapshots.
Catalog indexes, codex references, source artifacts, scripts, and this published site. The flat-file database lives here.
Legal Matters registry, Stonewall Archive database, All Email surface, and Tasks backlog. Cross-database case relations and morning briefs.
Personal Stonewall mirror for staging. Work OneDrive (read-only) for firm matter archives. PDFs, email exports, and screenshots flow from here into the repo.
FastAPI-powered case dashboard with search, deadlines, DG workflow readiness, and dossier views. Static export available for public-safe deployment.