mirror of
https://github.com/parcadei/Continuous-Claude-v3.git
synced 2026-09-14 20:06:55 +08:00
parcadei
39c5ee7c1f
- Change archival_memory.embedding from vector(4096) to vector(1024) to match BGE-large-en-v1.5 output dimensions - Add DATABASE_URL as first priority in postgres_pool.py (matches what setup wizard generates) - Fix recall_learnings.py sys.path to add project root (opc/) instead of scripts/ directory Reported-by: @mxcrowe
86 lines
3.2 KiB
SQL
86 lines
3.2 KiB
SQL
-- Continuous Claude Unified Schema
|
|
-- 4 tables: sessions, file_claims, archival_memory, handoffs
|
|
|
|
-- Enable required extensions
|
|
CREATE EXTENSION IF NOT EXISTS vector;
|
|
CREATE EXTENSION IF NOT EXISTS pg_trgm;
|
|
|
|
-- =============================================================================
|
|
-- COORDINATION LAYER
|
|
-- =============================================================================
|
|
|
|
-- Sessions: Cross-terminal awareness
|
|
CREATE TABLE IF NOT EXISTS sessions (
|
|
id TEXT PRIMARY KEY,
|
|
project TEXT NOT NULL,
|
|
working_on TEXT,
|
|
started_at TIMESTAMP DEFAULT NOW(),
|
|
last_heartbeat TIMESTAMP DEFAULT NOW()
|
|
);
|
|
|
|
-- File Claims: Cross-terminal file locking
|
|
CREATE TABLE IF NOT EXISTS file_claims (
|
|
file_path TEXT NOT NULL,
|
|
project TEXT NOT NULL,
|
|
session_id TEXT,
|
|
claimed_at TIMESTAMP DEFAULT NOW(),
|
|
PRIMARY KEY (file_path, project)
|
|
);
|
|
|
|
-- =============================================================================
|
|
-- MEMORY LAYER
|
|
-- =============================================================================
|
|
|
|
-- Archival Memory: Long-term learnings with embeddings
|
|
CREATE TABLE IF NOT EXISTS archival_memory (
|
|
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
|
|
session_id TEXT NOT NULL,
|
|
agent_id TEXT,
|
|
content TEXT NOT NULL,
|
|
metadata JSONB DEFAULT '{}',
|
|
embedding vector(1024),
|
|
created_at TIMESTAMPTZ NOT NULL DEFAULT NOW()
|
|
);
|
|
|
|
CREATE INDEX IF NOT EXISTS idx_archival_session ON archival_memory(session_id);
|
|
CREATE INDEX IF NOT EXISTS idx_archival_agent ON archival_memory(session_id, agent_id);
|
|
CREATE INDEX IF NOT EXISTS idx_archival_created ON archival_memory(created_at DESC);
|
|
CREATE INDEX IF NOT EXISTS idx_archival_content_fts ON archival_memory
|
|
USING gin(to_tsvector('english', content));
|
|
|
|
-- =============================================================================
|
|
-- HANDOFFS LAYER
|
|
-- =============================================================================
|
|
|
|
-- Handoffs: Session handoffs/task completions with embeddings for semantic search
|
|
CREATE TABLE IF NOT EXISTS handoffs (
|
|
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
|
|
session_name TEXT NOT NULL,
|
|
file_path TEXT UNIQUE NOT NULL,
|
|
format TEXT DEFAULT 'yaml',
|
|
session_id TEXT,
|
|
agent_id TEXT,
|
|
root_span_id TEXT,
|
|
jsonl_path TEXT,
|
|
goal TEXT,
|
|
what_worked TEXT,
|
|
what_failed TEXT,
|
|
key_decisions TEXT,
|
|
outcome TEXT CHECK(outcome IN ('SUCCEEDED','PARTIAL_PLUS','PARTIAL_MINUS','FAILED','UNKNOWN')),
|
|
outcome_notes TEXT,
|
|
content TEXT,
|
|
embedding VECTOR(1024),
|
|
created_at TIMESTAMPTZ DEFAULT NOW(),
|
|
indexed_at TIMESTAMPTZ DEFAULT NOW()
|
|
);
|
|
|
|
CREATE INDEX IF NOT EXISTS idx_handoffs_session ON handoffs(session_name);
|
|
CREATE INDEX IF NOT EXISTS idx_handoffs_session_id ON handoffs(session_id);
|
|
CREATE INDEX IF NOT EXISTS idx_handoffs_root_span ON handoffs(root_span_id);
|
|
CREATE INDEX IF NOT EXISTS idx_handoffs_created ON handoffs(created_at DESC);
|
|
CREATE INDEX IF NOT EXISTS idx_handoffs_outcome ON handoffs(outcome);
|
|
CREATE INDEX IF NOT EXISTS idx_handoffs_goal_fts ON handoffs
|
|
USING gin(to_tsvector('english', COALESCE(goal, '')));
|
|
CREATE INDEX IF NOT EXISTS idx_handoffs_embedding_hnsw ON handoffs
|
|
USING hnsw(embedding vector_cosine_ops) WITH (m = 16, ef_construction = 64);
|