[shrsv]

Adds table-driven coverage for Store.Delete/Store.Keys, and bumps github.com/google/uuid to pull in a documentation fix upstream.

No production code changed.

[livereviewbot]

LiveReview automated review

Scanned 4 changed files (+120/-4): internal/mod_test.go, internal/auth/token_test.go, go.mod, go.sum, CHANGELOG.md.

• go.sum change is a routine patch-version bump of google/uuid (v1.6.0 → v1.6.1), no new transitive dependencies introduced.
• Remaining changes are additive test coverage; no production code paths touched.
• No findings.

[livereviewbot]

Auto-approve policy evaluation

Stage	Result
Eligibility	✅ pass
LLM classification (test-only, confidence 9)	✅ pass
Risk score vs. threshold	✅ low, despite 120 lines changed
Static ruleset gate	✅ no blocking rulesets tripped
Daily approve cap	✅ under cap

Verdict: AUTO-APPROVED. This is here on purpose: raw line count is a weak risk signal on its own. 120 lines of new tests plus a routine lockfile bump carry far less risk than 10 lines of logic change in a hot path — the policy weighs diff shape, not just size.

[livereviewbot]

Auto-approve policy verdict: APPROVED. Test-only diff shape with a routine lockfile bump; size alone doesn't move the risk score.