<div class="product-slide dark"> <div class="hero"> <div> <img class="logo" src="assets/taranis-logo.svg" alt="Taranis AI" /> <div class="eyebrow" style="margin-top: 36px"> Open-source intelligence · AI-assisted analysis </div> <h1>Taranis AI</h1> <div class="tagline">From information overload<br />to intelligence that matters.</div> <p>Collect signals. Understand the story.<br />Deliver the briefing.</p> <div class="pills"> <span class="pill">Open source</span><span class="pill">Self-hosted</span ><span class="pill">Analyst in control</span> </div> </div> <div class="stack"> <div class="card"> <div class="eyebrow">The input</div> <h3>Feeds, articles, advisories, signals.</h3> </div> <div class="card blue"> <div class="eyebrow">The workspace</div> <h3>One connected workflow.<br />AI where it helps.</h3> </div> <div class="card"> <div class="eyebrow">The outcome</div> <h3>Reviewed intelligence.<br />Ready to share.</h3> </div> </div> </div> <div class="footer"> <span>taranis.ai</span><span>Product overview · September 2026 · 01</span> </div> </div> Note: Lead with the analyst's job: turn incoming information into something another person can act on. This deck follows the product workflow; it is not a project portfolio. Feature baseline: [release 1.5.0](https://github.com/taranis-ai/taranis-ai/releases/tag/1.5.0), published 16 September 2026, and the [product documentation](https://taranis.ai/_print/docs/), checked 21 September 2026. Optional integrations require configuration. --- <div class="product-slide"> <div class="eyebrow">The challenge</div> <h2>More information. The same analyst capacity.</h2> <p class="lede"> The hard part is deciding what deserves attention — and turning it into a useful briefing. </p> <div class="grid three"> <div class="card"> <h3>Repeated coverage</h3> <p>Several sources describe the same event. The analyst needs the shared story.</p> </div> <div class="card"> <h3>Scattered context</h3> <p>Names, indicators and earlier reporting sit across different articles.</p> </div> <div class="card"> <h3>A reporting deadline</h3> <p>Someone still has to select, explain, format and distribute the result.</p> </div> </div> <div class="takeaway">Taranis AI connects those steps in one working environment.</div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>02</span> </div> </div> Note: Use a simple scenario throughout: a new vulnerability appears in a vendor advisory, a news article and a community post. The analyst must decide whether it belongs in today’s briefing. This is an illustrative workflow, not a measured performance claim. --- <div class="product-slide dark"> <div class="eyebrow">The product in one view</div> <h2>Collect → Assess → Analyze → Publish</h2> <p class="lede">A continuous path from incoming material to a reviewed intelligence product.</p> <div class="grid four"> <div class="card"> <div class="number">01</div> <h3>Collect</h3> <p>Bring selected sources into the workspace.</p> </div> <div class="card"> <div class="number">02</div> <h3>Assess</h3> <p>Group coverage, review relevance and choose what matters.</p> </div> <div class="card"> <div class="number">03</div> <h3>Analyze</h3> <p>Combine stories with the analyst’s assessment.</p> </div> <div class="card"> <div class="number">04</div> <h3>Publish</h3> <p>Render the report and deliver it to its audience.</p> </div> </div> <div class="takeaway">News Item → Story → Report → Product</div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>03</span> </div> </div> Note: Explain the four objects before demonstrating the UI: a News Item is collected material; a Story groups related items; a Report contains the assessment; a Product is the rendered deliverable. Sources: [Assess](https://taranis.ai/docs/assess/), [Analyze](https://taranis.ai/docs/analyze/), [Publish](https://taranis.ai/docs/publish/). --- <div class="product-slide"> <div class="eyebrow">01 / Collect</div> <h2>Build the information stream you need.</h2> <div class="grid two"> <div class="stack"> <div class="card"> <h3>Web & RSS</h3> <p>Follow advisories, news sites and specialist publications.</p> </div> <div class="card"> <h3>Mastodon, MISP & Request Tracker</h3> <p>Bring social posts, shared intelligence and ticket context into the same workflow.</p> </div> </div> <div class="card blue"> <div class="eyebrow">Collection with control</div> <h3>Choose the sources.<br />Set the scope.</h3> <p>Scheduled collection and on-demand runs.</p> <p>Preview extraction before ingesting.</p> <p>Source groups and include/exclude word lists.</p> <p>Bulk setup and curated source lists.</p> </div> </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>04</span> </div> </div> Note: RSS and Simple Web support browser-based extraction when needed. Mastodon is a native collector in 1.5.0. Avoid claiming a universal social-media integration: the old Twitter and Email collectors were removed. Email remains a publishing channel. Sources: [Collectors](https://taranis.ai/docs/admin/collectors/), [1.5.0](https://github.com/taranis-ai/taranis-ai/releases/tag/1.5.0). --- <div class="product-slide"> <div class="eyebrow">02 / Assess</div> <h2>See the story behind the headlines.</h2> <div class="grid split"> <div class="stack"> <div class="card"> <h3>Bring coverage together</h3> <p>Group related items automatically or curate stories by hand.</p> </div> <div class="card"> <h3>Keep the useful context</h3> <p>Read the summary, inspect the original items and review extracted tags.</p> </div> <div class="card"> <h3>Make an explicit decision</h3> <p>Mark important, bookmark for follow-up or add to a report.</p> </div> </div> <div> <div class="image-panel"> <img src="assets/taranis_story.png" alt="Taranis Story view with collected news and analysis" /> </div> <div class="caption">Story view · existing product screenshot</div> </div> </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>05</span> </div> </div> Note: Clustering helps organize coverage; it does not establish that a source is correct. Analysts can inspect the items and adjust the story. Existing repository screenshots illustrate the workflow and may show an earlier UI. Source: [Assess](https://taranis.ai/docs/assess/). --- <div class="product-slide screenshot-slide"> <img src="assets/taranis_story.png" alt="Full-slide Taranis Story view with collected news and analysis" /> </div> Note: Full-slide view of the screenshot introduced on the previous slide. Walk through the source items, summary and extracted tags. This existing product screenshot may show an earlier UI. --- <div class="product-slide dark"> <div class="eyebrow">AI assistance / Available today</div> <h2>Less repetitive reading. More room for judgment.</h2> <div class="grid three"> <div class="card"> <h3>Summarize</h3> <p>Generate a concise story summary and title.</p> </div> <div class="card"> <h3>Extract entities</h3> <p>Identify people, organizations, places and other named entities.</p> </div> <div class="card"> <h3>Find indicators</h3> <p>Extract structured indicators for cybersecurity analysis.</p> </div> <div class="card"> <h3>Cluster coverage</h3> <p>Bring related material together for review.</p> </div> <div class="card"> <h3>Classify & tag</h3> <p>Apply cybersecurity classification, word lists and sentiment analysis.</p> </div> <div class="card blue"> <h3>Enrich indicators</h3> <p>Connect IntelOwl for additional IOC context.</p> </div> </div> <div class="takeaway">Enable the bots your workflow needs. Review the result.</div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>07</span> </div> </div> Note: Bot availability depends on configured services and endpoints. IntelOwl is optional, disabled by default and requires suitable analyzers; run IOC extraction first. Entity linking endpoints already exist in the optional LLM bot, so do not present all entity linking as future-only. Sources: [Bots](https://taranis.ai/docs/admin/bots/), [LLM Bot Service and IntelOwl sections](https://taranis.ai/_print/docs/). --- <div class="product-slide"> <div class="eyebrow">03 / Analyze</div> <h2>Turn selected stories into a briefing.</h2> <div class="grid two"> <div class="stack"> <div class="card"> <h3>A structure for your team</h3> <p>Configurable report types define the fields that matter.</p> </div> <div class="card"> <h3>Evidence plus assessment</h3> <p>Attach stories and write the analysis in the report editor.</p> </div> <div class="card"> <h3>A visible working history</h3> <p>Compare revisions and track whether a report is complete.</p> </div> </div> <div class="card blue"> <div class="eyebrow">Illustrative daily brief</div> <h3>What happened?<br />Why does it matter?<br />What should we watch?</h3> <div class="stack" style="margin-top: 34px"> <div class="flow-line" style="background: #164397; border-color: #4977cd"> Selected stories </div> <div class="flow-line" style="background: #164397; border-color: #4977cd"> Analyst assessment </div> <div class="flow-line" style="background: #164397; border-color: #4977cd"> Completed report </div> </div> </div> </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>08</span> </div> </div> Note: The questions illustrate a custom report structure, not fixed built-in fields or automatic AI report completion. Report Types are administrator-configured. Revision comparison is available; revision rollback is not. Source: [Analyze](https://taranis.ai/docs/analyze/). --- <div class="product-slide"> <div class="eyebrow">04 / Publish</div> <h2>Make intelligence useful beyond the workspace.</h2> <p class="lede"> Choose a template, render the product, review the output and send it where it belongs. </p> <div class="grid three"> <div class="card teal"> <h3>For people</h3> <p>PDF, HTML and plain-text briefings.</p> </div> <div class="card teal"> <h3>For systems</h3> <p>JSON and structured intelligence delivery.</p> </div> <div class="card blue"> <h3>For your workflow</h3> <p>Configured publishers and optional autopublish.</p> </div> </div> <div class="grid two"> <div class="card"> <h3>Delivery channels</h3> <p>Email, SFTP, S3-compatible storage and WordPress.</p> </div> <div class="card"> <h3>Intelligence exchange</h3> <p>MISP publishing and STIX bundles through TAXII 2.1.</p> </div> </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>09</span> </div> </div> Note: Outputs depend on Product Types, templates and configured publishers. Preview before delivery; autopublish requires a default publisher. TAXII requires a valid STIX bundle. Do not equate a Product count with successful publication. Sources: [Publish](https://taranis.ai/docs/publish/), [Publishers](https://taranis.ai/docs/admin/publishers/). --- <div class="product-slide"> <div class="eyebrow">Sharing / Available integration</div> <h2>Keep intelligence moving between teams.</h2> <div class="grid split"> <div class="stack"> <div class="card blue"> <div class="eyebrow">MISP story exchange · Experimental</div> <h3>Share context alongside indicators.</h3> <p>Push stories as MISP events and collect incoming updates.</p> </div> <div class="card"> <h3>Review differences</h3> <p>Inspect conflicting changes before resolving them.</p> </div> <div class="card"> <h3>Reduce manual updates</h3> <p>Optional auto-update for eligible stories and organization-owned events.</p> </div> </div> <div> <div class="image-panel"> <img src="assets/misp_taranis.png" alt="Illustration of Taranis AI and MISP integration" /> </div> <div class="caption"> Asynchronous exchange today · live collaboration is a future direction. </div> </div> </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>10</span> </div> </div> Note: Story exchange, auto-update and conflict resolution are experimental. Auto-update waits five minutes after the last change; proposals can block automatic updates. This is separate from publishing a rendered Product to MISP. Sources: [Connectors](https://taranis.ai/docs/admin/connectors/), [MISP Auto-Update](https://taranis.ai/_print/docs/). --- <div class="product-slide dark"> <div class="eyebrow">Deployment & control</div> <h2>Run the workspace on your terms.</h2> <div class="grid three"> <div class="card"> <h3>Self-hosted platform</h3> <p>Published container images for infrastructure you operate.</p> </div> <div class="card"> <h3>Configurable AI</h3> <p>Select compatible model endpoints and enable the capabilities you need.</p> </div> <div class="card"> <h3>Access controls</h3> <p>Roles, content ACLs and TLP visibility govern the analyst workspace.</p> </div> </div> <div class="grid two"> <div class="card"> <h3>Operational visibility</h3> <p>Health checks, job history and optional tracing help diagnose problems.</p> </div> <div class="card blue"> <h3>A modular foundation</h3> <p>Collectors, bots and publishers extend the workflow through defined interfaces.</p> </div> </div> <div class="takeaway"> Your deployment and provider choices determine where data is processed. </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>11</span> </div> </div> Note: Self-hosting the app does not automatically keep AI requests local: select and validate the provider accordingly. The current background stack is Redis/RQ, not Celery/RabbitMQ. Realtime notifications use Centrifugo in 1.5.0; they do not imply concurrent document editing. Deployment must replace default credentials and configure secrets before network exposure. Sources: [Deployment and Background Jobs](https://taranis.ai/_print/docs/), [1.5.0](https://github.com/taranis-ai/taranis-ai/releases/tag/1.5.0). --- <div class="product-slide"> <div class="eyebrow">One success story</div> <h2>An open product with growing momentum.</h2> <div class="grid three"> <div class="card"> <div class="eyebrow">Community</div> <div class="stat">1,218</div> <h3 style="margin-top: 22px">GitHub stars</h3> <p>179 forks · snapshot on 21 September 2026.</p> </div> <div class="card blue"> <div class="eyebrow">Operational context</div> <h3>CERT.at<br />& GovCERT</h3> <p>Supporting operational intelligence workflows.</p> <div style=" display: flex; gap: 24px; align-items: center; background: white; border-radius: 18px; padding: 20px; margin-top: 28px; " > <img src="assets/cert_at_logo.png" alt="CERT.at" style="height: 62px; width: auto" /><img src="assets/GovCERT_Logo_srgb.svg" alt="GovCERT" style="height: 62px; width: auto" /> </div> </div> <div class="card"> <div class="eyebrow">Open development</div> <div class="stat" style="font-size: 82px">EUPL</div> <h3 style="margin-top: 22px">Open source</h3> <p>Public code, releases and contributions. A platform teams can inspect and extend.</p> </div> </div> <div class="takeaway"> Community interest + operational feedback + continuous product development. </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>12</span> </div> </div> Note: Keep the organizational story on this single slide. Counts verified through the [GitHub repository API](https://api.github.com/repos/taranis-ai/taranis-ai) on 21 September 2026; refresh before presenting. Stars indicate interest, not deployment counts. CERT.at/GovCERT reference use comes from the supplied BMLV.md and strategy.md; no additional adoption numbers are inferred. License: EUPL-1.2. [Repository](https://github.com/taranis-ai/taranis-ai). --- <div class="product-slide dark"> <div class="eyebrow">Next / Planned direction</div> <h2>Live collaboration around a shared story.</h2> <div class="grid split"> <div class="stack"> <div class="card blue"> <div class="eyebrow">Future capability</div> <h3>A shared room for fast-moving situations.</h3> <p>Coordinate analysis while an incident, exercise or operation unfolds.</p> </div> <div class="card"> <h3>The intended experience</h3> <p>Shared story workspaces, visible analyst presence and coordinated updates.</p> </div> </div> <div> <div class="image-panel"> <img src="assets/taranis_live_collab_mockup.png" alt="Concept mockup of a future Taranis live collaboration room; not a released feature" /> </div> <div class="caption">CONCEPT MOCKUP · planned direction · no release date committed</div> </div> </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>13</span> </div> </div> Note: Live collaboration is the roadmap example requested for this presentation and appears in BMLV.md. Presence and shared story workspaces describe the intended experience, not an approved feature specification. Today’s Centrifugo updates and Chat are shipped; simultaneous collaborative editing is not claimed. Do not present the mockup as a screenshot of current functionality. --- <div class="product-slide"> <div class="eyebrow">Beyond / Product directions</div> <h2>More context. More media. Better assistance.</h2> <p class="lede"> Directions to explore as the platform evolves — beyond the capabilities already available. </p> <div class="grid three"> <div class="card"> <h3>Connected knowledge</h3> <p>Deeper use of linked entities and external knowledge across the analyst workflow.</p> </div> <div class="card"> <h3>More than text</h3> <p>Audio, video and image analysis as additional modular capabilities.</p> </div> <div class="card"> <h3>Assisted reporting</h3> <p>Help populate report structures and suggest relevant material for analyst review.</p> </div> </div> <div class="takeaway">Product direction · scope and timing to be validated</div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>14</span> </div> </div> Note: These directions synthesize the supplied BMLV.md and strategy.md, not commitments from the release notes. Basic entity-linking endpoints already exist in the optional LLM bot; this direction is deeper workflow integration. Chat is already shipped. Do not claim general multimodal ingestion, autonomous report generation or a full knowledge graph as current features. --- <div class="product-slide dark"> <div class="eyebrow">Live demo</div> <h2>Follow one story all the way through.</h2> <div class="grid two"> <div class="stack"> <div class="flow-line"><span>15</span>Open a focused Assess queue.</div> <div class="flow-line"><span>15</span>Inspect a story, its sources and AI tags.</div> <div class="flow-line"><span>15</span>Ask Chat to find related coverage.</div> <div class="flow-line"><span>15</span>Add the story to a report.</div> <div class="flow-line"><span>15</span>Render and preview the final product.</div> </div> <div class="card blue"> <div class="eyebrow">From information to intelligence</div> <h3 style="font-size: 48px"> The source stays visible.<br />The analyst adds judgment.<br />The result is ready to share. </h3> <div class="demo-link"><a href="https://taranis.ai/" style="color: white">taranis.ai</a></div> <div class="demo-link"> <a href="https://github.com/taranis-ai/taranis-ai" style="color: white" >github.com/taranis-ai/taranis-ai</a > </div> </div> </div> <div class="footer"> <span>Taranis AI · From information to intelligence</span><span>15</span> </div> </div> Note: Prepare a populated demo instance, a configured Chat provider and a working report/product template. Use an existing story if live collection is slow. End on the rendered preview; external publishing is a separate intentional action. If Chat is not enabled on the demonstration instance, explain the configuration prerequisite and show the other workflow steps. Documentation: [Taranis AI](https://taranis.ai/_print/docs/).