# Retrax Retrax is a standalone observability companion that can be attached to any application. Retrax lets user to: \- See every AI agent workflow run that your application executed \- Open a single run and trace it step by step — the execution graph, the time waterfall, and the call tree \- Inspect each step in detail: LLM calls, tool inputs/outputs, dependencies, tokens, and durations \- Debug failed runs, find slow steps, and audit exactly what the AI read and did This quick guide will help you get started with the application. # Getting Started **Preconditions:** - You have a Google account that is on the list of accounts allowed to access the dev environment. - You know the application URL: `https://dev.deconvolute.ai/Minion/retrax/#/` **Steps:** 1. Open the application URL in your browser. 2. If you do not have an active session, Retrax automatically redirects you to the Google sign-in page (`accounts.google.com`). 3. Enter the credentials of your allowed Google account and confirm. 4. After a successful sign-in you are returned to the application and the **Workflow runs** list opens. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783102278521.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783102278521.png) **Result:**You are authenticated and see the list of workflow runs. The session is remembered, so refreshing the page does not ask you to sign in again. # Interface Overview **There are two main screens:** ### 1. Workflow runs (Traces list) The landing screen. It lists every workflow run and lets you search and page through them. - **Header:** *Workflows / Workflow runs (Traces)* - **Filter field** — *"Filter this page…"*. - **Rows per page** selector — 20 / 40 / 80 / 120. - **Refresh** button and page navigation (*Previous / Next*). - **Run cards** — each shows the process name, status badge, executor duration, relative time, update date, author/owner and a unique ID. ### 2. Trace detail Opens when you click a run. It breaks one run down into nodes and steps. - **Top bar** - **View switcher:** **Graph / Waterfall / Tree**. - **Side panels group:** **Deps / Tool I/O / Details**. - **Visualization area** — the graph and waterfall. - **Right-hand panels** — *Inspector / LLM call / Trace tree*. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097248286.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097248286.png) **Result:**You understand where the run list lives, where a single trace is analysed, and which controls switch between views and panels. # Workflow runs **Preconditions:** - You are logged into Retrax and on the home screen This page lists every workflow run and lets you search and page through them. - **Header:** - **Filter field** - **Rows per page** selector — 20 / 40 / 80 / 120. - **Refresh** button and page navigation - **Run cards** — each shows the process name, status badge, executor , duration, relative time, update date, author/owner and a unique ID. User can perform several actions: ### Viewing the Runs List *What a run card shows:* - **Process name** The workflow that ran (e.g. *Project Chat*, *Document Ingestion*, *Template-Guided Analysis*). - **Status badge** (Completed, Running, Pending or Failed, each with its own icon/colour.) - **Description** Short text for some processes - **Executor** The agent / job / tool that performed the run. - **Duration** - **Time** Relative and absolute. - **Author / owner** Who created or owns the run. - **ID** The unique run identifier. **Notes:** - **Failed runs** show the error text right on the card, so you can triage without opening the trace. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097357477.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097357477.png) ### Filtering Runs **Steps:** 1. Click the filter field at the top of the list (*"Filter this page…"*). 2. Type any of the following to narrow the list: - **Process name** - **Executor** - **Author** - **ID fragment** 3. The list updates as you type, keeping only matching cards. 4. To clear the filter, empty the field — and the full list will be shown again. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097386598.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097386598.png) ### Pagination & Rows per Page #### Paging through runs 1. Use **Next** (*"Next workflow page"*) to move to the following page — the counter updates 2. Use **Previous** to go back one page. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097400013.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097400013.png) #### Rows per page 1. Open the **Rows per page** selector 2. Choose **20**, **40**, **80** or **120**. The default value is 40. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097421564.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097421564.png) #### Opening a Trace 1. Click any run card. 2. See the selected trace being opened. 3. User can also refresh the page by clicking \***Refresh** button [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097499969.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097499969.png) **Result:**User can successfully change the appearence of the list and move from the high-level list into the detailed breakdown of a single run. # Working with the Trace Each trace screen consists of several elements. **Layout of the trace screen:** - **Top bar** - **View switcher:** **Graph / Waterfall / Tree** - **Side panels group:** **Deps / Tool I/O / Details**. - **Visualization area** - **Right-hand panels** [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097577354.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097577354.png) [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097602873.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097602873.png) User can pick a view that fits his goal: - **Graph** - **Waterfall** - **Tree** #### Graph View A node graph of the run, with arrows connecting steps. It includes collapsible groups, guardrail badges, nested-step counters, and a final-result card. **User can perform several actions:** 1. **Zoom In / Out** — click the `+` / `−` buttons, or hold the mouse over the graph and scroll the wheel for smooth zoom. 2. **Fit View** — click *Fit View* to fit the whole graph into the visible area. 3. **Pan** — click and drag empty canvas to move around. 4. **Mini Map** — click a region of the mini map to jump there. 5. **Toggle Interactivity** (lock icon) — lock/unlock the canvas to prevent or allow moving node positions. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097716520.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097716520.png) #### Waterfall View **What it shows:**A time diagram of spans. Each row is a span with its name, type, duration and sometimes tokens. Groups are collapsible. Controls include **Critical path**, **Reset zoom**, a zoom-window slider, and hints *"⇧ drag to zoom · ⌘ wheel"*. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097829523.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097829523.png) #### Tree View A hierarchical *Trace tree*. Each node has a type icon, a name, an iteration label a duration, tokens (for LLM nodes) and tool-call counters. Branches are expandable. **User can perform several actions** 1. Open the **Tree** view to see the hierarchy from the root down. 2. Click a node's **expand arrow** to reveal its children; click again to collapse. 3. Read repeated steps by their numbering — numbering is sequential. 4. Read iteration labels — they follow the execution order. 5. Read per-node metrics on LLM nodes — duration and tokens. 6. Click any node to load its details into the **Inspector**; the selection syncs with Graph and Waterfall. 7. Expand the deepest branch to verify structure — indentation stays correct, with no truncation. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097856377.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097856377.png) **Result:**You have a complete, multi-angle picture of a single run and can drill into any step. # Side Panels ##### Inspector / Details The Inspector (opened via the **Details** side panel) shows everything about the currently selected node. It has a **Pretty / JSON** toggle and displays the node's type and status (e.g. *LLM\_CALL · COMPLETED*), name, an ownership line (*"in {parent} · Iter N · {duration}"*), a *Final results* block, and collapsible sections. User can close it with the **×**. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783097888076.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783097888076.png) The panel is read top-to-bottom. There are several sections: ### 1. Format toggle + identity header At the very top a **Pretty / JSON** toggle switches the whole panel between a human-readable layout (**Pretty**) and the raw payload (**JSON**). Under it is the node's identity: - **Type badge + status** — e.g. `LLM_CALL` with a green **COMPLETED** pill (other statuses: `RUNNING`, `PENDING`, `FAILED`). - **Name** — the step's name (e.g. *Thinking*). - **Ownership line** — The parent name is a link [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783099530090.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783099530090.png) ### 2. Final results The **Final results** block is the node's output. - **LLM node** — a quick **Open LLM call → Parsed messages** button that takes you to the full panel, plus short - **PROMPT** and **RESPONSE** previews. Each preview shows its size and has its own **Pretty / JSON** toggle. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783099606228.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783099606228.png) ### 3. Timing **Timing** shows exactly when the step was ran: - **Started** — absolute clock time plus a relative age - **Completed** — when it finished. - **Duration** — how long it took. - **Progress** — completion percentage (100% for a finished step). [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783100393258.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783100393258.png) ### 4. References **References** displays the identifiers that tie this node into the run: - **JOB ID** — this node's own id (the value used in `sel=`). - **PARENT JOB** — the id of the step that owns it. - **WORKFLOW** — the trace id it belongs to. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783100428328.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783100428328.png) ### 5. Tokens **Tokens** displays token breakdown that drives cost: - **Input tokens** / **Output tokens** — the split between prompt and completion. - **Total tokens** — their sum. - **LLM requests** — how many model calls this node made. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783100454998.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783100454998.png) ### 6. Metadata **Metadata** displays the low-level attributes of the step: - **State** - **Type** - **Class** - **Package** - **Owner** - **Iteration** - **Status message** [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783100504344.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783100504344.png) #### Tool I/O The **Tool I/O** panel shows the input and output of a tool call. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783100865031.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783100865031.png) ##### Header The top of the panel names the tool and how the call went: - the tool's **display name** - its **type and class** - a **status** icon and the call's **duration** on the right. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783101748801.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783101748801.png) ##### Input The **INPUT** block shows the parameters the tool was called with. Its header carries the payload **size** and a **Pretty / JSON** toggle, then each parameter is listed as a name/value pair. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783101803962.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783101803962.png) ##### Output The **OUTPUT** block shows the result. Its header also carries the **size** and a **Pretty / JSON** toggle. When the result is structured, Retrax lays it out as **collapsible sections** — so a big response stays readable instead of dumping a wall of JSON. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783101832753.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783101832753.png) **User can perform several types of actions:** 1. Select a **tool** node (e.g. *"Get Project Structure"*). The **Tool I/O** button becomes active. 2. Open **Tool I/O**. 3. Read the **input** — the parameters the tool was called with. 4. Read the **output** — the result the tool returned. 5. If a format toggle is available, switch between **Pretty** and **JSON**. ##### Dependencies (Deps) The **Dependencies** panel reveals scheduling dependencies between nodes. **User can perform several types of actions:** 1. Click **Deps** to open the dependencies panel / mode. 2. Find a node marked *"Has scheduling dependencies"* — the indicator confirms it has them. 3. Select a node that has dependencies — its related nodes / edges are highlighted. 4. Select an independent node — no dependencies are shown (empty), confirming it isn't waiting on anything. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783100886763.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783100886763.png) ##### LLM Call When user selects an LLM node, the middle side-panel button becomes **LLM call**. Opening it shows the details of that model call: - **model** and provider (e.g. *claude-sonnet-4-6 · BEDROCK*) - a timestamp - a status - a token breakdown - raw prompt/response. [![](https://wiki.slaq.ai/uploads/images/gallery/2026-07/scaled-1680-/image-1783100897269.png)](https://wiki.slaq.ai/uploads/images/gallery/2026-07/image-1783100897269.png)