Commit Graph

2159 Commits

Author SHA1 Message Date
DetroitTommy
1dbeb0827a added kilocode auth, needs adjusting 2026-02-15 13:44:26 -05:00
Luis Pater
d560c20c26 Merge branch 'router-for-me:main' into main v6.8.16-0 2026-02-15 14:49:13 +08:00
Luis Pater
5abeca1f9e Merge pull request #231 from ChrAlpha/main
feat(models): add Thinking support to GitHub Copilot models
v6.8.15-1
2026-02-15 14:48:04 +08:00
Luis Pater
294eac3a88 Merge branch 'main' into main 2026-02-15 14:47:52 +08:00
Luis Pater
a31104020c Merge pull request #230 from ultraplan-bit/main
fix(copilot): forward Claude-format tools to Copilot Responses API
2026-02-15 14:45:27 +08:00
Luis Pater
65bec4d734 Merge pull request #229 from Buywatermelon/fix/issue-222-kiro-alias-deletion
fix: preserve explicitly deleted kiro aliases across config reload
2026-02-15 14:42:42 +08:00
Luis Pater
edb2993838 Merge pull request #228 from xilu0/fix/antigravity-fetch-models-logging
fix(antigravity): add warn-level logging to silent failure paths in FetchAntigravityModels
2026-02-15 14:42:13 +08:00
Luis Pater
c0d8e0dec7 Merge pull request #226 from Skyuno/refactor/websearch-alignment
refactor(kiro): Kiro Web Search Logic & Executor Alignment
2026-02-15 14:41:46 +08:00
ChrAlpha
795da13d5d feat(tests): add comprehensive GitHub Copilot tests for reasoning effort levels 2026-02-15 06:40:52 +00:00
Luis Pater
55789df275 chore(docker): update Go base image to 1.26-alpine 2026-02-15 14:26:44 +08:00
ChrAlpha
9e652a3540 fix(github-copilot): remove 'xhigh' level from Thinking support 2026-02-15 06:12:08 +00:00
Luis Pater
46a6782065 refactor(all): replace manual pointer assignments with new to enhance code readability and maintainability 2026-02-15 14:10:10 +08:00
Luis Pater
c359f61859 fix(auth): normalize Gemini credential file prefix for consistency 2026-02-15 13:59:33 +08:00
Luis Pater
908c8eab5b Merge pull request #1543 from sususu98/feat/gemini-cli-google-one
feat(gemini-cli): add Google One login and improve auto-discovery
2026-02-15 13:58:21 +08:00
Luis Pater
f5f2c69233 Merge pull request #1595 from alexey-yanchenko/feature/cache-usage-from-codex-to-chat-completions
Pass cache usage from codex to openai chat completions
2026-02-15 13:56:46 +08:00
Alexey Yanchenko
63d4de5eea Pass cache usage from codex to openai chat completions 2026-02-15 12:04:15 +07:00
ChrAlpha
af15083496 feat(models): add Thinking support to GitHub Copilot models
Enhance the model definitions by introducing Thinking support with various levels for each model.
2026-02-15 03:16:08 +00:00
ultraplan-bit
c4722e42b1 fix(copilot): forward Claude-format tools to Copilot Responses API
The normalizeGitHubCopilotResponsesTools filter required type="function",
which dropped Claude-format tools (no type field, uses input_schema).
Relax the filter to accept tools without a type field and map input_schema
to parameters so tools are correctly sent to the upstream API.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-14 21:58:15 +08:00
Dave
f9a991365f Update internal/runtime/executor/antigravity_executor.go
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-02-14 10:56:36 +08:00
y
6df16bedba fix: preserve explicitly deleted kiro aliases across config reload (#222)
The delete handler now sets the channel value to nil instead of removing
the map key, and the sanitization loop preserves nil/empty channel entries
as 'disabled' markers.  This prevents SanitizeOAuthModelAlias from
re-injecting default kiro aliases after a user explicitly deletes them
through the management API.
2026-02-14 09:40:05 +08:00
Skyuno
632a2fd2f2 refactor: align GenerateSearchIndicatorEvents return type with other event builders
Change GenerateSearchIndicatorEvents to return [][]byte instead of []sseEvent
for consistency with BuildFallbackTextEvents and other event building functions.

Benefits:
- Consistent API across all event generation functions
- Eliminates intermediate sseEvent type conversion in caller
- Simplifies usage by returning ready-to-send SSE byte slices

This addresses the code quality feedback from PR #226 review.
2026-02-13 22:04:09 +08:00
Skyuno
5626637fbd security: remove query content from web search logs to prevent PII leakage
- Remove search query from iteration logs (Info level)
- Remove query and toolUseId from analysis logs (Info level)
- Remove query from non-stream result logs (Info level)
- Remove query from tool injection logs (Info level)
- Remove query from tool_use detection logs (Debug level)

This addresses the security concern raised in PR #226 review about
potential PII exposure in search query logs.
2026-02-13 22:04:09 +08:00
Skyuno
2db89211a9 kiro: use payloadRequestedModel for response model name
Align Kiro executor with all other executors (Claude, Gemini, OpenAI,
etc.) by using payloadRequestedModel(opts, req.Model) instead of
req.Model when constructing response model names.

This ensures model aliases are correctly reflected in responses:
- Execute: BuildClaudeResponse + TranslateNonStream
- ExecuteStream: streamToChannel
- handleWebSearchStream: BuildClaudeMessageStartEvent
- handleWebSearch: via executeNonStreamFallback (automatic)

Previously Kiro was the only executor using req.Model directly,
which exposed internal routed names instead of the user's alias.
2026-02-13 22:04:09 +08:00
Skyuno
587371eb14 refactor: align web search with executor layer patterns
Consolidate web search handler, SSE event generation, stream analysis,
and MCP HTTP I/O into the executor layer. Merge the separate
kiro_websearch_handler.go back into kiro_executor.go to align with
the single-file-per-executor convention. Translator retains only pure
data types, detection, and payload transformation.

Key changes:
- Move SSE construction (search indicators, fallback text, message_start)
  from translator to executor, consistent with streamToChannel pattern
- Move MCP handler (callMcpAPI, setMcpHeaders, fetchToolDescription)
  from translator to executor alongside other HTTP I/O
- Reuse applyDynamicFingerprint for MCP UA headers (eliminate duplication)
- Centralize MCP endpoint URL via BuildMcpEndpoint in translator
- Add atomic Set/GetWebSearchDescription for cross-layer tool desc cache
- Thread context.Context through MCP HTTP calls for cancellation support
- Thread usage reporter through all web search API call paths
- Add token expiry pre-check before MCP/GAR calls
- Clean up dead code (GenerateMessageID, webSearchAuthContext fp logic,
  ContainsWebSearchTool, StripWebSearchTool)
2026-02-13 22:04:09 +08:00
xiluo
75818b1e25 fix(antigravity): add warn-level logging to silent failure paths in FetchAntigravityModels
Add log.Warnf calls to all 7 silent return nil paths so operators can
diagnose why specific antigravity accounts fail to fetch models and get
unregistered without any log trail.

Covers: token errors, request creation failures, context cancellation,
network errors (after exhausting fallback URLs), body read errors,
unexpected HTTP status codes, and missing models field in response.
2026-02-13 18:01:46 +08:00
Luis Pater
cbe56955a9 Merge pull request #227 from router-for-me/plus
v6.8.15
v6.8.15-0
2026-02-13 12:50:52 +08:00
Luis Pater
8ea6ac913d Merge branch 'main' into plus 2026-02-13 12:50:39 +08:00
Luis Pater
ae1e8a5191 chore(runtime, registry): update Codex client version and GPT-5.3 model creation date 2026-02-13 12:47:48 +08:00
Luis Pater
b3ccc55f09 Merge pull request #1574 from fbettag/feat/gpt-5.3-codex-spark
feat(registry): add gpt-5.3-codex-spark model definition
2026-02-13 12:46:08 +08:00
Franz Bettag
1ce56d7413 Update internal/registry/model_definitions_static_data.go
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-02-12 23:37:27 +01:00
Franz Bettag
41a78be3a2 feat(registry): add gpt-5.3-codex-spark model definition 2026-02-12 23:24:08 +01:00
Luis Pater
1ff5de9a31 docs(readme): add CLIProxyAPI Dashboard to project list 2026-02-13 00:40:39 +08:00
Luis Pater
46a6853046 Merge pull request #1568 from itsmylife44/add-cliproxyapi-dashboard
Add CLIProxyAPI Dashboard to 'Who is with us?' section
2026-02-13 00:37:41 +08:00
xSpaM
4b2d40bd67 Add CLIProxyAPI Dashboard to 'Who is with us?' section 2026-02-12 17:15:46 +01:00
Luis Pater
726f1a590c Merge branch 'router-for-me:main' into main v6.8.14-0 2026-02-12 22:43:44 +08:00
Luis Pater
575881cb59 feat(registry): add new model definition for MiniMax-M2.5 2026-02-12 22:43:01 +08:00
Luis Pater
d02df0141b Merge pull request #224 from Buywatermelon/fix/kiro-assistant-first-message
fix(kiro): prepend placeholder user message when conversation starts with assistant role
v6.8.13-1
2026-02-12 15:11:10 +08:00
Luis Pater
e4bc9da913 Merge pull request #220 from jellyfish-p/main
fix(kiro): 修复之前提交的错误的application/cbor请求处理逻辑
2026-02-12 15:10:42 +08:00
Luis Pater
8c6be49625 Merge pull request #218 from ClubWeGo/fix/merge-assistant-tool-calls
fix: prevent merging assistant messages with tool_calls
2026-02-12 15:10:00 +08:00
Luis Pater
c727e4251f ci(github): trigger Docker image workflow on version tags matching v* 2026-02-12 15:09:16 +08:00
Luis Pater
99266be998 Merge pull request #216 from starsdream666/main
增加kiro新模型并根据其他提供商同模型配置Thinking
2026-02-12 15:08:37 +08:00
Luis Pater
d0f3fd96f8 Merge pull request #225 from router-for-me/main
v6.8.13
v6.8.13-0
2026-02-12 15:06:32 +08:00
hkfires
f361b2716d feat(registry): add glm-5 model to iflow 2026-02-12 11:13:28 +08:00
y
086d8d0d0b fix(kiro): prepend placeholder user message when conversation starts with assistant role
Kiro/AmazonQ API requires the conversation history to start with a user message.
Some clients (e.g., OpenClaw) send conversations starting with an assistant message,
which is valid for the native Claude API but causes 'Improperly formed request' (400)
on the Kiro endpoint.

This fix detects when the first message has role=assistant and prepends a minimal
placeholder user message ('.') to satisfy the Kiro API's message ordering requirement.

Upstream error: {"message":"Improperly formed request.","reason":null}
Verified: original request returns 400, fixed request returns 200.
2026-02-12 11:09:47 +08:00
jellyfish-p
627dee1dac fix(kiro): 修复之前提交的错误的application/cbor请求处理逻辑 2026-02-12 09:57:34 +08:00
Darley
55c3197fb8 fix(kiro): merge adjacent assistant messages while preserving tool_calls 2026-02-12 07:30:36 +08:00
Darley
5a2cf0d53c fix: prevent merging assistant messages with tool_calls
Adjacent assistant messages where any message contains tool_calls
were being merged by MergeAdjacentMessages, causing tool_calls to
be silently dropped. This led to orphaned tool results that could
not match any toolUse in history, resulting in Kiro API returning
'Improperly formed request.'

Now assistant messages with tool_calls are kept separate during
merge, preserving the tool call chain integrity.
2026-02-12 01:53:40 +08:00
starsdream666
2573358173 根据其他提供商同模型配置Thinking 2026-02-12 00:41:13 +08:00
starsdream666
09cd3cff91 增加kiro新模型:deepseek-3.2,minimax-m2.1,qwen3-coder-next,gpt-4o,gpt-4,gpt-4-turbo,gpt-3.5-turbo 2026-02-12 00:35:24 +08:00
starsdream666
ab0bf1b517 Merge branch 'router-for-me:main' into main 2026-02-11 16:20:20 +00:00