Coralogix & Skyflow team up on secure observability

Coralogix and Skyflow have partnered to protect sensitive customer data in observability logs, with a focus on debugging, incident response, security analysis and AI-related workflows.

The arrangement is designed to help organisations keep logs searchable and useful while limiting exposure of personal or regulated data that can appear in both structured fields and unstructured text.

Observability tools sit at the centre of software operations and security because they collect the logs, metrics and traces used to investigate system issues and threats. But those records can also contain customer identifiers and other sensitive information, creating tension between operational access and privacy controls.

Many vendors address that problem through redaction or masking. Coralogix and Skyflow argue that those methods can make data harder to search, correlate and analyse, especially when teams need to trace an incident across multiple events or let AI systems work with telemetry.

Under the partnership, Skyflow's technology replaces sensitive values with consistent tokens rather than removing them entirely. This keeps logs queryable while placing the underlying data under central access controls and audit rules.

Privacy model

The approach is aimed at organisations that need to investigate software failures or security incidents without broadly exposing customer data to engineers, analysts or downstream systems. It also targets businesses using AI tools on operational data, where raw information in logs might otherwise be ingested into dashboards, assistants or automated workflows.

Skyflow Chief Executive Anshu Sharma said the challenge is not only data exposure, but also the loss of operational context when data is stripped out.

"The traditional approach of redaction creates a false trade-off between safety and usefulness," said Anshu Sharma, CEO of Skyflow. "Once sensitive data is stripped out, teams lose the ability to search effectively, investigate incidents, or let AI agents reason over what actually happened. As a Runtime AI Data Control Platform, Skyflow ensures sensitive customer data stays governed and isolated, while observability data remains fully usable."

Coralogix frames the partnership as part of a broader shift in observability, where logs are no longer used only by engineers but increasingly feed automated systems and AI agents. That raises the stakes for data governance once information enters operational pipelines.

"Coralogix customers rely on observability data as a trusted system of record-supporting engineers, security teams, and the growing demands of AI-driven automation. They shouldn't have to choose between safeguarding sensitive customer data and maintaining operational efficiency. By partnering with Skyflow, we ensure they can achieve both seamlessly," said Ariel Assaraf, Chief Executive of Coralogix.

Residency issues

The companies also tied the partnership to data residency and sovereignty requirements. Coralogix said its platform already lets customers deploy observability workloads in specific geographic regions, while Skyflow's controls add policy-based governance over the sensitive data referenced by those workloads.

That is particularly relevant for businesses operating across multiple jurisdictions, especially in sectors such as finance, healthcare, retail and travel, where cross-border movement of customer data may face tighter scrutiny. In those environments, logs can become a hidden route for regulated information to spread through internal systems.

By keeping sensitive values isolated and allowing access only under policy, organisations can reduce that spread while maintaining visibility into incidents and application performance.

AI pressure

The partnership reflects a broader industry push to adapt observability and security tools for AI use. As companies deploy AI agents and assistants to examine telemetry, triage incidents and support operations teams, the quality and accessibility of log data become more important. At the same time, customer data in those records creates compliance and governance risks.

Coralogix said the joint model is designed to keep sensitive data out of logs, dashboards and downstream tools while preserving search, filtering and correlation across events. Approved workflows could also rehydrate data under policy when needed.

For customers, the commercial appeal will likely depend on whether tokenisation preserves enough context for investigations without adding operational complexity. Coralogix and Skyflow are betting that privacy controls built into observability pipelines will become more important as AI systems take on a larger role in software operations.

Rather than treating privacy and observability as opposing priorities, the partnership is built on the idea that sensitive customer data should stay governed and isolated while observability data remains usable for both people and AI systems.