1 item
Bridges the active *.darwinium.com tab to a local MCP server for external LLM clients (Claude Desktop, Claude Code). Darwinium Portal MCP connects an AI assistant to the Darwinium portal you already have open in your browser, so you can investigate traffic and author rules by asking for what you want in plain language. ABOUT DARWINIUM Darwinium is a digital security and fraud prevention platform. It instruments a customer journey — registration, login, checkout, payment, account change — at the perimeter, profiling each step for device, network, behavioural and identity signals. Those signals feed the features, models and rules that decide, event by event, whether to accept, challenge or reject. Fraud teams use the Darwinium portal to investigate what actually happened on live traffic, and to write the rules that decide what happens next. WHAT THIS EXTENSION DOES The extension is half of a two-part bridge. It reads the commands the portal page publishes and relays them to a local MCP (Model Context Protocol) server that your assistant runs on the same machine. The assistant can then see the page you are looking at and operate it on your behalf. FOR FRAUD ANALYSTS - Query live event data in Investigations using Darwinium query syntax, and get counts, aggregates, top-N breakdowns and quantile distributions back in the conversation. - Find the right attribute without memorising the schema. Search the attribute index in plain terms — "device identifier", "email", "geolocation" — and get the correct syntax, data type and example values. - Pull journey metadata and step names, browse and search the label catalogue, and apply labels to events. - Read dashboards and graphs, including screenshots, and have them interpreted against the underlying data rather than guessed at from a picture. FOR RULE AND SIGNAL AUTHORS - Draft signal conditions grounded in your workspace's real feature and signal names, which the assistant reads from the page instead of inventing. - Validate a proposed condition and count how many events it would actually match, before you commit to it. - Check workspace diagnostics after an edit, so a change is confirmed clean rather than assumed clean. - Move between Investigations, Workflows, Labels and Dashboards without losing the thread of the conversation.
Sep 3, 2026
rating_count is the Chrome Web Store ratings count, not a written-review count.
Media assets
Screenshots and videos on the listing.
Has promo video
Whether the listing includes at least one video.
Languages
Declared language locales.
Developer website
Listing exposes a developer website URL.
Contact email
Listing exposes a contact email.
Keyword in name
Case-insensitive substring match in the name.
Keyword in description
Case-insensitive substring match in the description.
Keyword occurrences in description
Count of case-insensitive occurrences in the description.
Category user-count percentile
Share of same-category extensions with fewer users (null if unknown).
These are transparent listing completeness / keyword signals, not a prediction of Chrome Web Store search ranking.