fix(test): shape mock config to prevent 500 in slot create test

[bolichen97]

Closing as duplicate of #6533 which applies the same fix with full local verification.

[github-actions]

Opus 4.8 Review — ✅ no blocking findings

Reviewed 210fa9966da55bc2411396bfac73641e95ed0aa8 — this comment is updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 210fa99

Verdict parsed from the review's SHA-scoped output markers for commit 210fa9966da55bc2411396bfac73641e95ed0aa8.

False positive or not applicable? A repository writer can comment:
/ai-review override fable 210fa9966da55bc2411396bfac73641e95ed0aa8: <one-sentence reason>

[github-actions]

GPT 5.6 Review — ✅ no blocking findings

GPT 5.6 completed its review of 210fa9966da55bc2411396bfac73641e95ed0aa8 and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 210fa99

False positive or not applicable? A repository writer can comment:
/ai-review override gpt 210fa9966da55bc2411396bfac73641e95ed0aa8: <one-sentence reason>

[github-actions]

Design Review (Fable 5) — ✅ PASS

Design-level review of 210fa9966da55bc2411396bfac73641e95ed0aa8 — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

Design-Verdict: PASS

Two-line mock shaping matches the handler's real falsy-config semantics; verified against the cfg.default_agent / resolve_agent_bindings paths in chat_handlers.py.

Suggestions

• The bare MagicMock() config remains a trap: any future config attribute the handler reads is a truthy Mock and re-breaks this test the same way. Building the mock from a real default KiroCrewConfig (overriding only dashboard.default_project) would fix the class, not the instance.

[DESIGN-REVIEWED] 210fa99