# Release Notes
### Latest highlights
{% columns %}
{% column width="50%" %}
{% hint style="success" icon="calendar-clock" %}
#### Scheduled Runs
Set repeat runs from **Target page → Scheduled Jobs**. Use daily, weekly, or custom cron.
{% endhint %}
{% endcolumn %}
{% column width="50%" %}
{% hint style="warning" icon="wand-magic-sparkles" %}
#### LLM-Driven Exploration
Add product context from **Product page → Edit**. Explorations choose next actions more reliably.
{% endhint %}
{% endcolumn %}
{% endcolumns %}
{% columns %}
{% column %}
{% hint style="danger" icon="vial" %}
#### Test Cases
Use **Sidebar → Tests** to run all tests or rerun one. Cases persist and stay manageable from the dashboard.
{% endhint %}
{% endcolumn %}
{% column %}
{% hint style="info" icon="key" %}
#### Credential Sets
Add sets from **Product page → Credential Sets**. Save users for different environments, then choose one at run time.
{% endhint %}
{% endcolumn %}
{% endcolumns %}
### Recent releases
| Date | Release | Notes |
| ---------- | -------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| 2026-04-27 | :wand-magic-sparkles: **LLM-Driven Exploration** | Add product context from **Product page → Edit** so explorations know what your site does and what to prioritise. The LLM now chooses actions more reliably across modals, overlays, and complex forms. |
| 2026-04-24 | :screwdriver-wrench: **Run Failure Diagnostics** | Failed runs now classify root cause more clearly. Platform issues and target issues are separated for faster triage. |
| 2026-04-14 | :key: **Credential Sets** | Add multiple credential sets from **Product page → Credential Sets → Add Credentials**. Choose the right set at run time for different users or environments. |
| 2026-04-08 | :vial: **Test Cases** | Use **Sidebar → Tests** to **Run all** or rerun a single case. Test cases auto-generate from explorations, persist for repeat execution, and can be deleted or managed from the dashboard. |
| 2026-04-07 | :diagram-project: **Cleaner Graph and Issue Views** | Use **Run detail → graph viewer** to **Expand graph** or **Collapse graph**. Similar pages now collapse together, and duplicate issues are merged into one unique item. |
| 2026-04-06 | :brain: **State Detection Upgrades** | State detection identifies page changes more accurately with structural, semantic, and network signals. |
| 2026-04-06 | :tag: **Test Run Badges** | Status badges reflect partial failures more accurately. |
| 2026-04-01 | :calendar-clock: **Scheduled Runs** | Set repeat runs from the **Scheduled Jobs** section at the bottom of a target page. Use daily, weekly, or custom cron schedules, manage them from the new Schedules page, view per-schedule history, and trigger any schedule immediately. |
| 2026-03-27 | :arrows-rotate: **Incremental Graph Updates** | Target graphs now update incrementally after each exploration batch. |
| 2026-03-26 | :tower-broadcast: **Live Stream Preview** | Watch explorations and test runs in real time from the dashboard. |
| 2026-03-26 | :chart-line: **Run Metrics** | Exploration and test run metrics are now persisted for review. |
| 2026-03-20 | :circle-play: **Test Run Video Playback** | Test run detail pages now include video recordings for easier debugging. |
| 2026-03-20 | :right-to-bracket: **Improved Login Detection** | Login and authentication flow detection is now more reliable. |
| 2026-03-19 | :bug: **Failure Diagnostics** | Test runs now include clearer debugging context, structured summaries, and side-by-side path comparison. |
| 2026-03-16 | :eye: **Improved State Detection** | Accessibility-tree-based state detection now filters out visual layout noise. |
| 2026-03-13 | :filter: **Pagination and Filtering** | The Runs page now supports server-side pagination, sorting, search, and filtering. |
| 2026-03-12 | :users: **Organisations and Invitations** | Invite team members, assign roles, and restrict API key management to org admins and above. |
| 2026-03-12 | :signal: **Live Run Status** | Running explorations now show animated status indicators. |
| 2026-03-10 | :share-nodes: **Target Graph** | A persistent unified graph now aggregates exploration results across multiple batches. |
| 2026-03-09 | :bullseye: **Richer Goal Descriptions** | Auto-generated goals now include more contextual detail from the explored application. |
| 2026-03-05 | :lock: **Progressive Login Support** | Credential injection now handles multi-step login flows. |
| 2026-03-05 | :rocket: **Simpler Exploration Start** | URL is now optional when starting an exploration. |
Older releases
| Date | Release | Notes |
| ---------- | ------------------------------------------------------------------------------------ | ------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| 2026-02-28 | :list-check: **Test Runs** | Full test run management now includes listing, filtering, sorting, detail views, action timelines, and per-goal test triggers. |
| 2026-02-23 | :route: **Guided Exploration** | Path-based guided runs now replay discovered routes through your application for verification. |
| 2026-02-20 | :network-wired: **Network Activity Enrichment** | Network capture now includes full response data, richer edge details, and automatic goal generation at the end of each exploration. |
| 2026-02-19 | :building: **Organisation Management** | Added organisation management tools in the dashboard. |
| 2026-02-16 | :key: **Saved Credentials** | Save login credentials per target with encryption at rest. Goals now appear directly on graph nodes and detail panels. |
| 2026-02-13 | :wave-square: **Network Activity Analysis** | Explorations now capture and analyse network activity, including API calls and form submissions, to improve state detection and goal identification. |
| 2026-02-11 | :database: **Persistent Exploration Results** | Exploration results are now stored persistently, merge automatically when runs complete, and improve support for SPAs and iFrames. |
| 2026-02-06 | :sparkles: **AI-Powered Goal Generation** | Duku now generates natural-language descriptions of discovered user flows automatically. Interactive graph views are also available on target and run detail pages. |
| 2026-02-04 | :compass-drafting: **New Exploration Strategies** | New exploration strategies improve page discovery and coverage. Custom input values and more exploration parameters are also available in the dashboard. |
---
# Agent Instructions: 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:
```
GET https://docs.duku.ai/changelog/release-notes.md?ask=
```
The question should be specific, self-contained, and written in natural language.
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.