Turn CodeQL's new AI prompt-injection detection into a stable repository security contract with positive, negative, and SARIF assertions.
GitHub announced on July 10, 2026 that CodeQL 2.26.0 adds AI prompt-injection detection. Enabling a query is useful; owning a regression test is better.
Primary source: GitHub Changelog, July 10, 2026.
The examples below are implementation templates, not results from a repository I tested.
Model a path, not a phrase
A useful fixture contains an untrusted source, prompt construction, and a model sink:
Most production AI features are still glueing prompts together with string concatenation in some random Node service. The prompt…
Sentinel is a prompt regression gate. It runs an eval suite against both versions of a changed...
Prompt injection isn't a bug you patch — it's structural. Why filters and stronger prompts fail, and the architectural defenses…
injection-arena is a 10-level prompt-injection challenge platform with a server-side judge, canary tokens, and a fully offline…
If you're building AI applications, you've probably heard about prompt injection attacks. Users...
I've been building Agenci — open-source CI/CD testing for AI agents (functional tests, security...
