> For the complete documentation index, see [llms.txt](https://intuitem.gitbook.io/ciso-assistant/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://intuitem.gitbook.io/ciso-assistant/concepts/specialised-modules/object-classification.md). # Object classifications Organisation-defined sensitivity schemes — like TLP — with ordered, coloured levels An **object classification** is a scheme that marks how sensitive an object is — think **TLP** (Traffic Light Protocol), or a custom scale aligned to a customer's or an authority's own tiers. Each scheme is an ordered set of **levels**, and any object that supports classification can be tagged with one of them. CISO Assistant ships with **TLP** built in, and lets you tune it or define your own schemes alongside it. ## Why it exists Organisations exchange documents with customers, partners, and authorities that each impose their own sensitivity tiers. One party works in TLP; another in a national CONFIDENTIAL / SECRET scale; a customer in its own bespoke levels. Object classifications let you define those schemes once and apply them consistently, with colour coding people already recognise. ## Mental model ```mermaid graph LR Scheme[Object classification] -->|comprises| Level[Classification level] Level -. classifies .-> Doc[Document] ``` A **scheme comprises an ordered set of levels**. Each level carries a **rank** (its position in the order, from least to most sensitive), a **colour**, an **abbreviation**, and a translatable name and description. A level can then **classify** an object — today, a **document** — which surfaces it as a coloured badge and, on published PDFs, a page marking. The dashed edge is deliberate: classifying an object is always optional. ## How it works A scheme is a thin container with a name and an **ID** (`TLP`, for example). Its meaning lives in its **levels**, which are ordered by **rank**: * **Rank** sets the order and carries semantic weight — a higher rank is more sensitive. TLP runs `CLEAR < GREEN < AMBER < AMBER+STRICT < RED`. * **Colour** is the swatch shown wherever the level appears; the platform picks readable text automatically over it. * **Abbreviation** is the short token on the badge (`AMBER`, `RED`). * **Name** and **description** are translatable, so a level reads in the user's language. The built-in **TLP** scheme follows the FIRST TLP 2.0 standard — five levels with their official colours. Its structure is protected: you can hide levels you don't use, but you can't delete or reorder the canonical set. ## Ordering carries meaning Because levels are ranked rather than just listed, the order is data, not decoration — "at least AMBER" is a well-defined comparison. Today a classification is a **visual marking**; the ordinal rank is the foundation for comparing and, in future, gating access by level. {% hint style="info" %} Classification is currently cosmetic — a badge and a document marking. The rank is designed so that a later access-control layer can treat a clearance as granting every lower level, without reworking the data. {% endhint %} ## Customising a scheme Open a scheme to reach its level editor, where you can: * **Add** a custom level with **Add level** — set its abbreviation, name, and colour; it slots in at the end of the order. * **Reorder** levels with the up/down controls (rank follows the order). * **Hide** a level with its visibility toggle — it stops being offered without being destroyed. * **Edit** or **delete** custom levels. Built-in levels (and the built-in TLP scheme itself) can only be **hidden**, never deleted or restructured — so the canonical TLP always stays intact, and a hidden level stays hidden across restarts. ## Applying a classification to a document When you create or edit a document, the **Classification** field lets you pick a level. Once set, the level shows as a coloured badge on the document catalog, in the reader, and in the documents table, and it is stamped on every page of the document's exported PDF (for example `TLP:AMBER`, in the level's colour). Deleting a level a document points at simply clears the field — the document is never removed. Documents are the first object type to carry a classification; the same mechanism extends to other objects over time. ## Scoping Schemes and their levels live in the global (root) folder — they apply organisation-wide. Multiple schemes can coexist, so TLP and a customer-specific scheme are both available at once. ## Where you find it In the sidebar under **Extra → Object classifications**. The list shows every scheme; open one to manage its levels. Access is governed by the standard permissions — viewing is available to every role, while creating and editing schemes and levels is reserved for administrators. ## Related * [Documents](/ciso-assistant/concepts/governance/documents.md) — the first object type that carries a classification * [Terminology](/ciso-assistant/concepts/specialised-modules/terminology.md) — a related organisation-defined override layer --- # 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://intuitem.gitbook.io/ciso-assistant/concepts/specialised-modules/object-classification.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.