- Add settings profiles to zsh env completion (was proxy-only)
- Document intentional ANTHROPIC_MODEL omission in transformToOpenAI
- Use getCcsDir() in error hint instead of hardcoded ~/.ccs/
- Export parseFlag and add 5 unit tests for flag parsing
- Add fish and PowerShell single-quote escaping tests
- Validate env var keys match ^[A-Za-z_][A-Za-z0-9_]*$ before output
to prevent injection via crafted config files
- Add env command to all 4 shell completion scripts (bash, zsh, fish,
PowerShell) with sub-completions for --format and --shell flags
Switch formatExportLine from double quotes to single quotes to prevent
shell metacharacter expansion ($(), backticks, etc.) when output is
consumed via eval. Also fix parseFlag to handle values containing =,
remove unused test imports, and add empty-output guard after format
transformation.
New `ccs env <profile>` command exports shell-evaluable environment
variables for OpenCode, Cursor, Continue, and other third-party tools.
Supports --format (openai|anthropic|raw) and --shell (auto|bash|fish|
powershell) flags. Auto-detects shell from $SHELL env var.
Closes#503
Replace permissive JSON Schema Draft-07 whitelist with strict
Gemini-compatible field set (22 fields). The "examples" field
in Claude Code tool schemas caused 400 errors from Gemini API.
Also strips other unsupported fields: $ref, $defs, oneOf, allOf,
additionalProperties, const, if/then/else, etc.
Safe change — sanitizer only runs for CLIProxy profiles (Gemini,
Codex, Antigravity), never for direct Anthropic API requests.
Closes#155
Consolidate 5 duplicate boolean tracking variables into a lifecycle
object with an update() method, reducing event detection from 4
locations to centralized calls. Add pause/drain/resume backpressure
to the passthrough path to prevent unbounded memory buffering with
slow clients.
Add message_start/message_delta/message_stop detection to the
remaining-buffer block in the SSE processing path for defensive
completeness. Addresses code review feedback on PR #493.
Track hasReceivedMessageDelta and hasReceivedMessageStop in both
streaming paths (pipe-through and SSE-processing). Conditionally
omit these events from buildSyntheticErrorResponse() when upstream
already sent them, preventing protocol violations.
Closes#491
* fix(cliproxy): disable 1M extended context for gemini-claude-opus-4-6-thinking
Antigravity backend only supports 256k context window for this model,
not 1M as previously declared. Set extendedContext: false with TODO
comment for easy re-enable when backend adds support.
Closes#490
* fix(cliproxy): strip [1m] suffix at runtime for models without extended context
Users with existing [1m] in saved agy.settings.json will now have it
automatically stripped at runtime when the model no longer supports
extended context. Prevents misleading context window claims.
* fix(cliproxy): guard against empty upstream SSE responses in agy profile
When CLIProxyAPIPlus drops unsigned thinking blocks during sub-agent
execution, the response stream can contain no content_block_start or
message_delta events. This causes Claude Code CLI to crash with
"No assistant message found".
Add empty response detection in the tool sanitization proxy's streaming
handler. Both the pipe-through and SSE-processing paths now track
whether meaningful content was received. If upstream sent data but no
content blocks on a 200 OK, a synthetic minimal valid SSE response is
injected to prevent the client crash and surface a clear error message.
Closes#350
* fix(cliproxy): improve empty response detection and add tests
Address code review findings:
- Remove message_delta from content detection (lifecycle event, not
content); only content_block_start indicates actual content
- Add try-catch in end handlers to handle client disconnects gracefully
- Add 3 integration tests: empty stream injection, normal stream
passthrough, 4xx/5xx non-injection
* fix(cliproxy): avoid duplicate message_start in synthetic response
Track whether upstream already sent a message_start event. When
injecting the synthetic error response, omit message_start if upstream
already sent one, preventing duplicate events in the SSE stream.
Addresses PR review feedback from ccs-reviewer[bot].
* test(cliproxy): add SSE processing path and real failure mode tests
Address code review observations:
- Add test exercising SSE processing path (sanitized tool names) for
empty response detection, ensuring both code paths are covered
- Add test mirroring real failure mode where upstream sends only
message_start then ends abruptly (no message_delta/message_stop)
- Document duplicate message_start assumption with inline comment
CLIProxyAPIPlus already handles background token refresh for kiro (every
1 min via background_refresh.go) and other providers. CCS was incorrectly
returning "not implemented" errors for these providers, causing noisy
worker logs and preventing proper token expiry tracking.
Changes:
- Classify providers into CCS-managed (gemini), CLIProxy-delegated
(codex, agy, kiro, ghcp, qwen, iflow), and not-implemented (claude)
- Delegated providers return success with delegated flag instead of error
- Skip refresh_token requirement for delegated providers in expiry checker
- Filter delegated providers from background worker refresh loop
Closes#487
- update default model in code-reviewer.ts
- update REVIEW_MODEL, ANTHROPIC_MODEL, ANTHROPIC_DEFAULT_OPUS_MODEL
in ai-review.yml workflow
- aligns reviewer with AGY model catalog default
When CCS launches Claude with CLAUDE_CONFIG_DIR pointing to the isolated
instance path, agent team teammates spawned via tmux split-window don't
inherit it because tmux creates new panes from its own session environment.
This causes teammates to use ~/.claude/ instead of ~/.ccs/instances/{profile}/,
creating a split-brain in shared state (tasks, teams, mailbox).
Sets key CCS env vars in the tmux session environment via `tmux setenv` so
all new panes inherit the correct config directory.
* fix(version): show active config path instead of deprecated config.json
The version command was using deprecated getConfigPath() which always
returned config.json path. Now uses getActiveConfigPath() which shows
config.yaml in unified mode or config.json in legacy mode.
* chore(release): 7.37.1-dev.1 [skip ci]
* fix(ui): use native dynamic import to fix Node 24 ESM/CJS interop
TypeScript compiles import() to require() when targeting CommonJS,
which breaks ESM packages like ora on Node 24. Use new Function()
to create native dynamic import at runtime, bypassing TS transform.
Closes#472
* chore(release): 7.37.1-dev.2 [skip ci]
* fix(env): strip ANTHROPIC_* from account/default profiles
Account and default profiles inherit process.env which may contain
stale ANTHROPIC_BASE_URL from prior CLIProxy sessions. This causes
ConnectionRefused errors when Claude tries to hit an unavailable proxy.
Settings-based profiles already handle this by explicitly injecting
their own ANTHROPIC_* values. This fix applies the same protection
to account/default profiles by stripping ANTHROPIC_* before spawn.
Closes#474
* test(env): add unit tests for stripAnthropicEnv
Address code review feedback from PR #475. Tests cover:
- Removing all ANTHROPIC_* keys
- Preserving non-ANTHROPIC keys
- Empty object handling
- Undefined value preservation
- Case sensitivity (only uppercase ANTHROPIC_)
- All ANTHROPIC_ prefixed variants stripped
* chore(release): 7.37.1-dev.3 [skip ci]
* feat(cliproxy): add extended context support for 1M token window
Add --1m and --no-1m flags to enable/disable 1M token context window.
Uses Claude Code's [1m] suffix mechanism.
Behavior:
- Gemini models: auto-enabled by default
- Claude models: opt-in with --1m flag
- New extendedContext field in model catalog
Also adds Claude Opus 4.6 to model catalog with extended context support.
Closes#103
* feat(ui): add extended context toggle in dashboard model config
- Add ExtendedContextToggle component for 1M token context window
- Add Claude Opus 4.6 (claude-opus-4-6-20260203) to model catalogs
- Mark Gemini and Claude models with extendedContext: true
- Toggle only appears when selected model supports extended context
- Auto-enabled info for native Gemini, opt-in info for Claude
Part of extended context feature implementation for issue #103.
* fix: address code review findings and CI failure
- Fix CI error: add missing 'provider' prop to ModelConfigSection
- Fix case sensitivity in applyExtendedContextSuffix
- Fix whitespace handling in stripModelSuffixes
- Handle --1m=value and --no-1m=value CLI patterns
- Add warning when --1m used on unsupported model
- Sync agy catalog: add extendedContext to gemini-3-pro-preview
- Extract isNativeGeminiModel to shared utility (DRY)
- Add 21 unit tests for extended-context-config
* fix(catalog): rename claude-opus-4-6-20260203 to claude-opus-4-6
* fix(ui): wire extended context toggle through component tree
- Add extendedContextEnabled and toggleExtendedContext to useProviderEditor hook
- Store setting as CCS_EXTENDED_CONTEXT env var in provider settings
- Pass props through provider-editor → model-config-tab → model-config-section
- Update UseProviderEditorReturn type with new properties
* fix(ui): apply [1m] suffix directly to model strings in settings
- Toggle now applies/strips [1m] suffix to all ANTHROPIC_*MODEL env vars
- Extended context detected by checking if any model has [1m] suffix
- Remove legacy CCS_EXTENDED_CONTEXT flag approach
- Add suffix utilities: applyExtendedContextSuffix, stripExtendedContextSuffix
- Raw Configuration now shows actual model values with [1m] suffix
* fix(qwen): update model catalog with correct context windows and tier mappings
- Update context window specs from official Alibaba docs:
- Qwen3 Coder Plus: 1M context (was 32K)
- Qwen3 Max: 256K context (flagship)
- Qwen3 Coder Flash: fast code generation
- Fix preset mappings for Claude tier equivalence:
- Opus → qwen3-max (flagship 256K)
- Sonnet → qwen3-coder-plus (balanced 1M)
- Haiku → qwen3-coder-flash (fast)
- Update provider descriptions to reflect 256K-1M context range
- Add all 7 Qwen models from CLIProxyAPI: qwen3-coder-plus, qwen3-max,
qwen3-max-preview, qwen3-235b, qwen3-vl-plus, qwen3-coder-flash, qwen3-32b
Closes#478
* fix(ui): only apply [1m] suffix to ANTHROPIC_MODEL, fix toggle refresh
- Only ANTHROPIC_MODEL gets [1m] suffix, not tier mappings
- Strip [1m] when looking up model in catalog to prevent toggle disappearing
- Fix odd page refresh when toggling extended context
* chore(release): 7.37.1-dev.4 [skip ci]
* fix(ui): remove duplicate import in use-provider-editor
* chore: address PR review feedback - sync comment and unused import
* chore(release): 7.37.1-dev.5 [skip ci]
* feat(cliproxy): add Opus 4.6 to Antigravity model catalog (#482)
* feat(cliproxy): add Opus 4.6 to Antigravity model catalog
- Add gemini-claude-opus-4-6-thinking as new default agy model
- Update preset mappings to route opus tier to Opus 4.6
- Bump CLIProxy fallback versions to v6.8.2
- Keep Opus 4.5 as previous flagship option
* fix(cliproxy): include oauth-model-alias in config generation
Root cause: CLIProxy config.yaml was missing Opus 4.6 alias because:
1. CLIProxyPlus startup migration is disabled (intentional)
2. CCS config generator never wrote oauth-model-alias section
3. Existing users with outdated aliases got 502 on Opus 4.6
Fix:
- Add DEFAULT_ANTIGRAVITY_ALIASES to config generator
- Generate oauth-model-alias section in config.yaml template
- Preserve claude-api-key and custom aliases during regeneration
- Bump config version to v6 to trigger auto-regeneration
- Update model catalog tests for new model count
* fix(cliproxy): preserve YAML indentation in extractYamlSection
- Replace .trim() with regex to strip only leading/trailing newlines,
preserving 2-space indent on claude-api-key children
- Skip standalone comments at col 0 in section boundary detection
- Update stale test description (4 → 5 models)
* chore(release): 7.37.1-dev.6 [skip ci]
---------
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
For Device Code OAuth providers (ghcp, qwen), switch from /start-url
to /start endpoint which spawns CLIProxyAPI binary and emits WebSocket
events with userCode. DeviceCodeDialog then displays the code properly.
Closes#460
Tam Nhu Tran