> For the complete documentation index, see [llms.txt](https://t-matic.gitbook.io/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://t-matic.gitbook.io/docs/getting-started/core-concepts.md). # Core Concepts T-Matic has its own vocabulary for the things it does. This page defines the key terms so the rest of the documentation makes sense. If something in the docs is unclear, this is the first place to check. *** ## Workspace A workspace is the top-level container for your organization in T-Matic. Everything belongs to a workspace: projects, members, billing, and settings. One organization = one workspace. All members of a workspace share access to all projects within it. Billing (subscription and token credits) is managed at the workspace level, not per user. *** ## Project A project is a container for a single brand you are analyzing. When you create a project, you provide the brand's website URL. T-Matic crawls the site, builds a Brand DNA profile, and populates the project's knowledge base from what it finds. Projects are independent of each other. You can have one project per client brand, or multiple projects comparing different brands. Members can access all projects in the workspace. *** ## Brand DNA Brand DNA is T-Matic's structured AI-generated profile of a brand. It is the output of the analysis pipeline that runs when you create a project. Brand DNA captures: * **Identity** — brand name, tagline, overview, core values * **Audience** — who the brand is targeting, buyer personas * **Voice & tone** — how the brand communicates (formal/casual, authoritative/empathetic, etc.) * **Visual identity** — color palette, typography, aesthetic style (extracted from screenshots) * **Messaging** — key value propositions, proof points, calls to action Brand DNA is generated in two phases. The first draft is ready within 2–4 minutes from the homepage alone. The final version is enriched after T-Matic crawls the rest of the site's key pages. You will see the status transition from **Draft ready** to **Complete** when the enrichment finishes. *** ## Source A source is an individual web page that T-Matic has analyzed. Every URL that T-Matic crawls becomes a source in the project. Sources are not just raw page data. Each source is processed through a multi-pass AI pipeline that extracts: * **Factual content** — entities, claims, value propositions, calls to action * **Voice analysis** — tone, linguistic patterns, persuasion techniques, audience signals * **Knowledge graph data** — relationship triples (e.g., `Brand OFFERS Product`, `Product SOLVES PainPoint`) * **Taxonomy** — content type, buyer journey stage, messaging gaps Sources feed the Brand DNA and populate the Knowledge Graph. The more sources a project has, the richer its intelligence becomes. *** ## Connector A connector is an integration that links an external platform to T-Matic. Currently available connectors: * **Website crawler** — the primary connector, activated when you provide a URL * **Meta / Instagram** — imports published social content and enables post scheduling The term "connector" refers to the integration mechanism. A "source" is the specific piece of data that comes through a connector. One Instagram connector can produce many sources (individual posts, profile data). *** ## Knowledge Graph The Knowledge Graph is a structured map of every entity and relationship T-Matic has extracted across all sources in a project. Entities are typed nodes: brands, products, audiences, pain points, competitors, platforms, certifications, value propositions, and proof points. Relationships are typed edges with a controlled vocabulary: `OFFERS`, `SOLVES`, `TARGETS`, `COMPETES_WITH`, `INTEGRATES_WITH`, `CERTIFIED_BY`, `DEPLOYS_ON`, `CLAIMS`, `DIFFERENTIATES_FROM`, and more. Every relationship carries a confidence score (0–1) and is grounded in an evidence excerpt from the source page. This means you can trace every claim in the graph back to the specific text that produced it. The graph is built up incrementally as sources are added. Entities discovered across multiple pages are merged — if three pages all mention the same product, they contribute to one product node with richer relationships. *** ## Maven Maven is T-Matic's AI assistant. It has access to everything in the project's knowledge base: the Brand DNA, all source analyses, and the knowledge graph. You interact with Maven through natural language. Ask it questions about the brand, request comparisons, ask for content ideas, or query specific claims. Maven's answers are grounded in the brand's actual website content, not generic AI knowledge. Each project has its own Maven conversation. The conversation history persists so you can build on previous questions. *** ## Pipeline The pipeline is the automated sequence of steps that runs whenever T-Matic analyzes content. You will hear this word in the context of both: * **Brand DNA pipeline** — the crawl-and-analyze sequence that runs when a project is created * **Source analysis pipeline** — the per-page AI extraction pipeline (factual → voice → graph → taxonomy → validate) Pipeline status is visible in the UI. If a pipeline fails, the status shows **Failed** and you can view the error detail. Pipelines can be re-triggered manually from the project or source view. *** ## Lineage Lineage refers to the chain of evidence connecting a claim, entity, or insight back to its source. Every fact in the Brand DNA and Knowledge Graph can be traced to the specific source URL and text excerpt that produced it. Lineage matters because it lets you verify AI output. If Maven tells you a brand claims to be "the market leader in enterprise data security," lineage tells you exactly which page said it and what the surrounding context was. *** ## Report vs. Dashboard **Dashboard** is the live, auto-updating view of a project's current state — pipeline status, recent activity, Brand DNA overview. It reflects real-time data. **Reports** (coming soon) are point-in-time snapshots of a project's analysis, formatted for export or sharing. A report captures the Brand DNA, key sources, graph highlights, and Maven conversation excerpts as of a specific date. --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://t-matic.gitbook.io/docs/getting-started/core-concepts.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.