Knowledge Bank

Knowledge Bank adds the runbooks, heuristics, and judgment that live only in your team’s head. This includes understanding the unwritten rules of an architecture, the business significance of a particular service, or the specific debugging workflow that has been refined over years of practice.

Why Knowledge Bank?

Even the best AI can miss critical context that only lives in engineers’ heads:

• Which services are business-critical

• Which alerts are usually noise

• Where your team actually starts when debugging an issue

Knowledge Bank closes this gap by letting you encode that company-specific operational and tribal knowledge directly into the investigation process: not replacing Traversal’s intelligence, but sharpening it with your experience.

What you can upload to your Knowledge Bank

Customers can use Knowledge Bank to upload any specific pieces of tribal knowledge, instructions, or whatever they want the agent to know about their system.

By default, we recommend starting with runbooks and then also giving us context on your system. Some examples of knowledge items might be:

• Prioritization rules “Staging alerts are usually low priority unless near a release.”

• Debugging workflows “For database issues, start with connection pool metrics before logs.”

• External dependencies “Payments alerts often correlate with PSP outages—check their status page first.”

• Service context “This cluster supports checkout and is customer-facing.”

How to use Knowledge Bank

Knowledge Bank provides three distinct channels for capturing knowledge: manual input, in-session feedback, and automated learning.

Step 1. Upload Runbooks and Custom Context

Your team has documentation that would help any investigation. Now you can give it directly to Traversal: upload a PDF, paste text, or write instructions in a simple form.

Each knowledge item has:

• Title (required)

• Instruction (required, free-form text)

• Use when (optional, applicability hint)

The most effective types of knowledge items are runbooks. We always recommend starting with uploading these.

Step 2. Automated Learning (Implicit Learning)

Traversal has always learned implicitly from how teams interact with it. Knowledge Bank makes that implicit learning visible and controllable.

Traversal reflects on any follow up question or message you send to Traversal. So even if you aren't explicitly trying to teach Traversal, it is always learning from you. For example, when you redirect an investigation—such as moving from logs to traces or metrics—Traversal learns from those choices automatically, without requiring intentional feedback. After an investigation concludes:

• Traversal extracts customer-specific insights

• Generic or obvious observations are filtered out

• New memories appear in your Memory Bank for review

These learned insights appear in your Knowledge Bank, where you can edit, approve, or delete anything it learns.

Step 3. In-Session Feedback (Explicit Teaching)

After any investigation, you can explicitly tell Traversal what it got right, what it missed, and what should happen differently next time. This feedback is intentional, user-directed, and specific to the investigation that just occurred, giving Traversal clear guidance on how to adjust future investigations.

Get Started

After logging onto Traversal, go to Settings → Knowledge Bank to upload your first knowledge item.