jiantw83
0b59ca6023
refactor(LLM CLI): 明確化 CLI 參數與驗證輸出
2026-06-29 10:49:22 +00:00
jiantw83
68ce92b0e5
fix(TLS 驗證): 移除 git 與 OpenRouter 的不安全憑證略過設定
2026-06-29 10:49:22 +00:00
Jeffery
9ec30e1abb
chore(config): 將 codex 預設模型更新為 gpt-5.4-mini
CI / Release Tag Version (pull_request) Successful in 5s
CI / AI Code Review (pull_request) Successful in 6m6s
2026-06-29 16:01:38 +08:00
Jeffery
34a9eb6d4e
fix(llm): 移除 codex exec 已不支援的 --ask-for-approval 參數
2026-06-29 16:01:35 +08:00
jiantw83
6f997083ef
feat(ai-code-review): 支援 Antigravity CLI
CI / AI Code Review (pull_request) Failing after 17s
CI / Release Tag Version (pull_request) Successful in 4s
2026-06-27 13:47:44 +00:00
jiantw83
4fcb240208
test(ai-code-review): 更新 CLI 呼叫與前置驗證測試
2026-06-27 13:30:45 +00:00
jiantw83
08a72a8c7f
feat(ai-code-review): 改用 AI 助理 CLI 執行審查
2026-06-27 13:30:40 +00:00
jiantw83
93be261b90
fix(OpenCode 呼叫): 降低並行請求並重試暫時性錯誤
CI / AI Code Review (pull_request) Failing after 3s
2026-06-26 09:36:33 +00:00
jiantw83
6c3e7b9d37
fix(LLM 審查流程): 避免單一角色失敗中止 pipeline
CI / AI Code Review (pull_request) Failing after 7s
2026-06-26 09:26:12 +00:00
jiantw83
2fe03e94f0
test(TLS 驗證): 補上憑證忽略行為測試
2026-06-26 09:20:57 +00:00
jiantw83
bc0df06445
fix(TLS 驗證): 套用全域憑證忽略設定
2026-06-26 09:20:57 +00:00
Jeffery and Claude Opus 4.8
f0a42cd98d
test(git): 補 commitAndPush 在 push 失敗時不中斷流程的測試
...
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 15:26:08 +08:00
Jeffery and Claude Opus 4.8
998b5ca5ac
perf(config/findings): getOpenCodeHttpsAgent 改模組單例、normalizeText 加 memoization 快取
...
另含 comments.js bySeverity 改用預建 LEVEL_RANK Map（隨 fix commit 一併提交）。
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 15:26:08 +08:00
Jeffery and Claude Opus 4.8
402630fa69
fix(json/comments): validateJSONArrayFile 先驗證再寫入避免毀損原檔、findingRow 對空值防呆
...
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 15:26:08 +08:00
Jeffery and Claude Opus 4.8
55b5c6f68e
docs(comments): 移除與 isUnclassified JSDoc 重複的冗餘行內註解
...
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 14:41:41 +08:00
Jeffery and Claude Opus 4.8
775cc575ae
test(app): 補齊 isSafeRepoPath/extractBalancedJSON/normalizeText/repairJSONArrayWithAI/格式化函式測試
...
為可測性 export 三個內部函式（isSafeRepoPath、extractBalancedJSON/extractJSONText、normalizeText）。
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 14:41:41 +08:00
Jeffery and Claude Opus 4.8
fc10ab7266
fix(git): bot commit 改用 comment token (PAT) 推送以重新觸發 workflow
...
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 14:41:41 +08:00
Jeffery and Claude Opus 4.8
3ced8fa688
chore(app): 新增 .gitignore 忽略 node_modules
...
CI / AI Code Review (pull_request) Failing after 1m45s
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 14:16:04 +08:00
Jeffery and Claude Opus 4.8
1378f03595
docs(ai-code-review): 補齊各模組 JSDoc、指令檔逐行註解並重建 README
...
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 14:16:04 +08:00
Jeffery and Claude Opus 4.8
303104bb20
test(test 目錄): 將 12 個單元測試移至 app/test 並更新 npm test 指令
...
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com >
2026-06-26 14:15:55 +08:00
jiantw83
8272166fde
fix(review publish): 降級處理 Gitea review 批次發布失敗
2026-06-25 15:26:09 +00:00
jiantw83
afec5a9387
fix(action image): 安裝 git 供前置遠端驗證使用
2026-06-25 15:17:51 +00:00
jiantw83
8315e1da2a
fix(gitea TLS): 固定略過憑證驗證並改用 Gitea context
2026-06-25 10:28:30 +00:00
jiantw83
65f27f544b
Refactor LLM verification and update dependencies
...
- Removed OpenAI and related dependencies from package.json and package-lock.json.
- Simplified LLM verification logic to focus on OpenCode server.
- Updated tests to reflect changes in LLM provider handling.
- Added CI and CD workflows for automated processes in Gitea.
2026-06-25 10:17:08 +00:00
jiantw83
525f6f9350
feat: add role management and usage tracking for AI code review
...
- Implemented role parsing and loading from markdown files, including attributes like name, side, focus, badge, color, and personality.
- Created functions to build prompts for analysis, line location, and verdicts based on roles.
- Added tests for role management functionalities to ensure correct parsing and loading of roles.
- Developed usage tracking for AI assistant interactions, including token usage and rate limits.
- Implemented functions to extract and record usage data from various LLM providers.
- Added tests for usage tracking functionalities to validate correct accumulation and reporting of usage statistics.
2026-06-25 09:34:59 +00:00
jiantw83
Jeffery