liveBook · Manning

Secret scanning

As explained in AI Governance

Secret scanning is the automated detection of credentials, keys, and other sensitive authentication material in source code and related artifacts.

Secret scanning detects exposed credentials and tokens. In GenAI deployments, extend it beyond code to prompt logs, training corpora, and RAG document stores; scan uploaded files before model processing and outputs before release. section 4.3.5

Make detection part of an operational control path:

  • Use tools such as GitGuardian or GitHub’s built-in scanner, and route detections to alerting or credential revocation. section 4.3.5
  • Send alerts for credentials found in public repositories to automated revocation; use rapid manual revocation for internal detections when automatic action could disrupt production. section 4.3.5

ClaimAssist CI pipeline

After a dependency incident exposed gaps in supply-chain visibility, the team added secret scanning and dependency scanning to CI and began maintaining a lightweight software bill of materials for AI-specific components. section 4.9

How AI Governance covers it

How deep it goes
Extends scanning to GenAI data paths and CI, and discusses alerting and revocation without detailing scanner internals.
Examples
Scans prompt logs, training corpora, RAG stores, uploads, outputs, and the ClaimAssist CI pipeline.
What you can do after
Apply secret scanning with GitGuardian or GitHub’s built-in scanner across code and GenAI-specific surfaces.
How it connects
Connects secret scanning to credential management, CI pipelines, SIEM alerting, and incident response.

Also explained in

Compare how 2 books explain secret scanning →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of AI Governance. The section links point to the parts of the book it draws on.