How Stratta works
Why Stratta gives you exact, verifiable citations instead of plausible-sounding guesses.
Stratta is designed around one goal: every answer must be traceable to the source document. When the result ends up under a signature, an answer you can't verify is worse than no answer. This page explains the approach in plain terms.
The problem with ordinary AI search
Most AI document tools chop a document into thousands of small fragments, turn each into a numeric vector ("embedding"), and retrieve the fragments that look mathematically similar to your question. This works for casual Q&A, but it has two failure modes that matter for norms, laws and regulations:
- Fuzzy provenance: A retrieved fragment may not map cleanly to a section number, an article or a page, so the citation is approximate or invented.
- Lost structure: These documents are deeply hierarchical and cross-referenced. Flat fragments lose the chapter → section → clause (or title → chapter → article) structure that tells you where a rule actually lives.
Stratta's approach: TreeRAG
Instead of embeddings, Stratta stores each document as its real table of contents: a tree of chapters, sections, sub-sections, and annexes. Your agent navigates that tree the way you flip through the printed document:
See what's available
The agent lists the documents in your workspace.
Open the table of contents
It reads the chapter structure, then drills into the relevant section.
Read the exact section
It opens the section's full content: text, formulas (in LaTeX), tables, and figures.
Follow the cross-references
A rule is often spread over several texts (for example the action in SIA 261 and the safety factors in SIA 260, or a law and its implementing ordinance). The agent follows those links before concluding.
Because each section keeps its exact path and page range, every citation the agent produces
([SIA 261, 8.2, p. 44]) points to a real, openable location. This is what we mean by
deterministic citations.
TreeRAG is inspired by the PageIndex approach to document reasoning. Read the deeper rationale in Concepts → TreeRAG.
The two halves of Stratta
Where you sign up, deposit your documents, create API keys, and manage your organization and your plan.
The MCP server you install in your agent. It authenticates with your API key and gives the agent the tools to read and ingest your documents.
There is no separate chat interface to learn. You use the agent you already have; Stratta simply adds your documents.
Your documents are yours
Documents are scoped to your organization. You deposit them yourself, and each time you declare that you hold the rights (a licence, the publisher's authorization, or a public-domain text such as a law or a published decision). They are visible only to your space, never a shared public corpus. This keeps you in control of licensing and confidentiality. See Concepts → Multi-tenant norms.