The response interceptor in client.ts transforms errors into plain
objects {status, code, message}, but catch blocks were checking
error.response?.status (AxiosError format) which never matched.
- Add error field passthrough in client.ts interceptor
- Refactor BulkEditAccountModal to use pre-check API (checkMixedChannelRisk)
before submit, matching the single edit flow
- Fix EditAccountModal catch blocks to use interceptor error format
- Add bulk-update mixed channel unit tests
- Move mixed channel check before any DB writes in BulkUpdateAccounts
- Return 409 from BulkUpdate handler for MixedChannelError
- Add ConfirmDialog to BulkEditAccountModal for mixed channel warning
- Update mixed channel warning message to Chinese
Apply Claude Max cache billing to usage before returning ForwardResult
in Antigravity Forward, ensuring RecordUsage gets the same simulated
usage that clients see. Restore apply+fallback in RecordUsage for
consistency across GatewayService and Antigravity paths.
Extract setupClaudeMaxStreamingHook and applyClaudeMaxNonStreamingRewrite
facade functions to helpers file. RecordUsage now uses detect-only (no
mutation), client response rewriting handled at Forward layer.
Previously 2K images used the same base price as 1K ($0.134).
Now 2K uses 1.5x multiplier ($0.201), consistent with 4K using 2x ($0.268).
- Backend: add 2K size branch in getDefaultImagePrice
- Frontend: update 2K placeholder from 0.134 to 0.201
- Tests: update assertions for new 2K default price
- Add migration 060 to update model_mapping for all antigravity accounts
- Remove gemini-3-pro-image and gemini-3-pro-image-preview mappings
- Add gemini-3.1-flash-image and gemini-3.1-flash-image-preview mappings
- Update frontend usage window to show GImage for new model
- Update isImageGenerationModel to support new model
- Add migration 060 to update model_mapping for all antigravity accounts
- Remove gemini-3-pro-image and gemini-3-pro-image-preview mappings
- Add gemini-3.1-flash-image and gemini-3.1-flash-image-preview mappings
- Update frontend usage window to show GImage for new model
- Update isImageGenerationModel to support new model
PR #635 returned HTTP 200 with {"input_tokens": 0} when upstream doesn't
support count_tokens (404). This caused Claude Code CLI to trust the zero
value, believing context uses 0 tokens, so auto-compression never triggers.
Fix: return 404 with proper error body so CLI falls back to its local
tokenizer for accurate estimation. Return nil (not error) to avoid
polluting ops error metrics with expected 404s.
Affected paths:
- Passthrough APIKey accounts: upstream 404 now passed through as 404
- Antigravity accounts: same fix (was also returning fake 200)
