Organizational Context

Klaudia uses three context layers to ground every investigation in your environment — not generic best practices.

Klaudia.md

A blueprint file written once by your team. It captures service dependencies, hard constraints, and rules that must never be violated. Klaudia.md is automatically loaded into every investigation session, ensuring remediations never violate your environment's specific constraints.

Getting Started

Read here on how to enable Klaudia.md files.

Knowledge Base

Connect your existing runbooks, postmortems, and troubleshooting guides. Klaudia semantically searches the knowledge base on-demand per incident, retrieving the most relevant runbook for the active failure pattern and applying your team's specific procedures rather than generic responses.

Getting Started

Read here on how to enable Klaudia.md files.

Klaudia Memory

Klaudia retains investigation history across sessions — what happened, what was tried, and what resolved each incident. Similar incidents are recognized instantly, and resolution playbooks are auto-indexed over time. Resolution speed improves with every incident.

Why it matters?

• Faster time-to-root-cause. Klaudia skips paths she has already walked in your environment. Investigations start with the likely answer already in hand.
• Accuracy that compounds. Every resolved incident strengthens Klaudia’s understanding of your stack — your services, your dependencies, your failure modes.
• Proven fixes, not guesses. When a remediation has worked before for a similar issue, Klaudia surfaces it with the context of what was done and why.
• Tribal knowledge, preserved. Patterns that normally live only in your senior SREs’ heads become part of the platform — and don’t leave when people do.
• Less noise. Klaudia learns which signals are benign in your environment and stops chasing them.

How it works

After every investigation, Klaudia distills what happened into structured memories — the kind a senior SRE would mentally file away for next time. These aren’t raw logs; they’re the useful lessons:

• Failure correlations — “when X times out, it’s usually Y.”
• Resolution playbooks — “rollback resolved this; restart did not.”
• Topology shortcuts — “this pod’s issues usually trace back to that certificate.”
• Temporal patterns — “this service OOMs every Monday after the batch job.”

When a new incident starts, Klaudia looks up what she already knows — by exact resource match and by similar situations on other resources — and uses it to investigate smarter. Patterns that recur gain confidence over time; patterns that stop recurring fade and are pruned.

Shared across your account

Memory is account-wide. A learning from one incident is available to every future investigation in your account — across clusters, namespaces, and teams. This cross-incident learning is what makes Memory valuable, and it is not configurable per user or team.

You're in control

• Account-level toggle — admins can turn Klaudia Memory on or off for the entire account at any time.
• Transparent by default — when Klaudia uses a memory during an investigation, it’s surfaced in the session so you can see exactly what she relied on.

The toggles control whether Klaudia reads from and writes to Memory. The account-wide scope itself — learning that crosses incidents — is part of how the capability works and isn’t a separate setting.