Public evidence snapshot
rr_security_prompt_boundaries
This build-owned snapshot keeps review provenance available to anonymous learners even when the original editorial repository or pull request is private or renamed.
- Course
openclaw-security-hardening- Lesson
openclaw-security-hardening.prompt-boundaries- Reviewer
- academy-security-source-review
- Method
- source-review
- Outcome
- approved
- Reviewed
- Content commit
159895e2b55dc872979109c1e1acea14fdeb0574- Dependency snapshot
0d5124f0d2ef5eb3ffd5fc55b5e068c412f1abd370ccb02d6708b37f7b2a2d57
Recorded sources
- openclaw-docs-index — docs-security-2026-07-30
Snapshot SHA-256:fe0abe5ec840c32ddc09ea6270c5ab216f2383b058c19fd7471caecf596d5d27
Recorded canonical reference:https://docs.openclaw.ai/gateway/security
Limitations
- Approval covers the bounded 2026.7.1 learning exercises and cited security guidance; operators must validate controls against their own deployment and use disposable credentials in labs.
Original review reference: https://github.com/openclaw-academy/learn/pull/20