Skip to content

getting-started.md

Latest commit

 

History

History
203 lines (145 loc) · 5.58 KB

getting-started.md

File metadata and controls

203 lines (145 loc) · 5.58 KB

Getting Started

Install qp-vault and have a governed knowledge store running in under 5 minutes.

Install

pip install qp-vault

Zero config. SQLite backend, trust-weighted search, content-addressed storage, Merkle verification, Membrane screening, and RBAC built in.

For additional features:

pip install qp-vault[sqlite]         # Async SQLite (aiosqlite)
pip install qp-vault[postgres]       # PostgreSQL + pgvector hybrid search
pip install qp-vault[encryption]     # AES-256-GCM encryption at rest
pip install qp-vault[pq]             # ML-KEM-768 + ML-DSA-65 post-quantum crypto
pip install qp-vault[capsule]        # Cryptographic audit trail via qp-capsule
pip install qp-vault[docling]        # 25+ document formats (PDF, DOCX, PPTX, etc.)
pip install qp-vault[local]          # Local embeddings (sentence-transformers, air-gap)
pip install qp-vault[openai]         # OpenAI embeddings (cloud)
pip install qp-vault[integrity]      # Near-duplicate + contradiction detection (numpy)
pip install qp-vault[fastapi]        # 22+ REST API endpoints
pip install qp-vault[cli]            # vault command-line tool (15 commands)
pip install qp-vault[all]            # Everything

Combine extras with commas to install exactly what you need:

# Production API server with PostgreSQL and encryption
pip install qp-vault[postgres,encryption,pq,capsule,fastapi]

# Air-gapped deployment with local embeddings and CLI
pip install qp-vault[local,encryption,cli]

# Development with document processing and integrity checks
pip install qp-vault[docling,integrity,cli]

Create a Vault

from qp_vault import Vault

vault = Vault("./my-knowledge")

Creates ./my-knowledge/ with SQLite database and audit log. No configuration required.

Add Resources

# From text
vault.add("Incident response: acknowledge within 15 minutes...",
          name="sop-incident.md", trust_tier="canonical")

# From a file
vault.add("path/to/report.pdf", trust_tier="working")

# With tenant isolation
vault.add("Tenant-specific content", tenant_id="site-123")

# Batch add
vault.add_batch(["doc1.md", "doc2.md", "doc3.md"], trust_tier="working")

Resources are automatically: chunked, hashed (SHA3-256), screened by Membrane, indexed, and audited.

Search

results = vault.search("incident response procedure")

for r in results:
    print(f"[{r.trust_tier.value}] {r.resource_name} (relevance={r.relevance:.3f})")

Results are deduplicated (one per resource), ranked by: (vector + text) * trust * freshness * layer_boost.

Multi-Keyword Grep

# Find documents where multiple concepts converge
results = vault.grep(["incident", "response", "P0", "escalation"])

for r in results:
    meta = r.explain_metadata
    print(f"{r.resource_name} — {len(meta['matched_keywords'])}/{4} keywords matched")
    print(f"  snippet: {meta['snippet']}")

Single-pass FTS5 query. Scored by keyword coverage (coord factor), text relevance, and term proximity. No embedder required.

Retrieve Content

text = vault.get_content(resource.id)
print(text)

Replace Content

old, new = vault.replace(resource.id, "Updated policy v2 content")
# old.lifecycle = "superseded", new = active version

Verify Integrity

result = vault.verify()           # Full vault Merkle tree
result = vault.verify(resource.id) # Single resource
proof = vault.export_proof(resource.id)  # For auditors

Health Score

score = vault.health()
print(f"Overall: {score.overall}/100")

# Per-resource health
score = vault.health(resource.id)

RBAC

# Reader: search and verify only
vault = Vault("./knowledge", role="reader")

# Writer: add and modify
vault = Vault("./knowledge", role="writer")

# Admin: full access
vault = Vault("./knowledge", role="admin")

Multi-Tenancy

# Per-call tenant isolation
vault.add("content", tenant_id="site-123")
vault.search("query", tenant_id="site-123")

# Or lock the vault to a single tenant
vault = Vault("./knowledge", tenant_id="site-123")

Knowledge Graph

Track entities, relationships, and mentions across your vault:

# Create entities
alice = vault.graph.create_node(name="Alice", entity_type="person")
acme = vault.graph.create_node(name="Acme Corp", entity_type="company")

# Connect them
vault.graph.create_edge(source_id=alice.id, target_id=acme.id, relation_type="works_at")

# Track mentions in documents
resource = vault.add("Alice leads engineering at Acme Corp.", name="team.md")
vault.graph.track_mention(alice.id, resource.id, context_snippet="Alice leads engineering")

# Search and traverse
results = vault.graph.search_nodes("Alice")
neighbors = vault.graph.neighbors(alice.id, depth=2)
backlinks = vault.graph.get_backlinks(alice.id)

Works on both PostgreSQL and SQLite. See Knowledge Graph for extraction, detection, and wikilinks.

CLI

pip install qp-vault[cli]

vault init ./knowledge
vault add report.pdf --trust canonical
vault search "revenue" --top-k 5
vault list --trust canonical
vault verify
vault health
vault status

15 commands. See CLI Reference.

Next Steps