Open source & contributions / case 05
Memtruth
Evidence and document runtime
An open-source SDK and runtime for evidence, parsing, corpus contracts, chunking, projection, diagnostics, provenance, and citation-aware AI applications.
- Classification
- Open Source
- Delivery status
- Public open-source repository
- Full record
- Open canonical case
The problem
AI applications often turn documents into untraceable text, making it difficult to inspect where a field came from, what parser produced it, or whether a result can be audited.
System arrangement
How the system works
Documents pass through versioned parsing and evidence contracts so every projection, retrieval result, and consuming application can retain provenance.
- 01SourceSource documents
Files keep original identity
- 02CapabilityParser contract
Versioned structured output
- 03ControlEvidence record
Field, source, and locator
- 04CapabilityChunk + projection
Retrieval-ready representations
- 05External systemConsuming application
Search, extraction, or agent
- 06OutcomeCitation + audit
Readable evidence trail
- 07ControlSchema policy
Compatibility across releases
My role
I shaped and contributed evidence, provenance, citation, parser, CLI, and release contracts across the public repository and its runtime integrations.
Public capabilities
- 01
Document parsing contracts
- 02
Per-field citation and provenance structures
- 03
Corpus and chunking contracts
- 04
Projection and diagnostics
- 05
Audit-oriented CLI behavior
Evidence
- •
Public repository includes explicit citation and provenance contracts.
- •
Tests cover evidence-aware extraction and readable audit output.
- •
Public releases preserve clone-safe onboarding and language-specific SDK structure.
Product decisions
- 01
Keep evidence explicit instead of burying source context in prompts.
- 02
Use schema-versioned public contracts for outputs that other systems consume.
- 03
Do not imply that a named trust document is automatically audit-grade.