diff --git a/pages/src/content/docs/en/overview.md b/pages/src/content/docs/en/overview.md index d47800f..ee1601c 100644 --- a/pages/src/content/docs/en/overview.md +++ b/pages/src/content/docs/en/overview.md @@ -80,57 +80,6 @@ The agent's strengths are concentrated where they matter most: result is a purpose-built set of [six tools](../tools/) that is more stable and predictable than a generic agent toolkit. -## How the pipeline fits together - -```mermaid -flowchart TD - Start["ocr review --from main --to feature"] - S1["1. Resolve LLM endpoint
config / env / shell rc"] - S2["2. Load diffs from git
workspace / commit / range"] - S3["3. Filter files
binary → user_exclude → user_include
→ ext allowlist → default path"] - S4["4. Drop diffs > 80% of MAX_TOKENS"] - S5["5. Dispatch per-file sub-agents (concurrent)

For each file:
  a. Plan phase (if changed lines ≥ 50)
  b. Main loop: LLM → tool calls → … → task_done
  c. code_comment results collected (async via worker pool)

Memory compression triggers when context
exceeds 60 % (async) or 80 % (sync) of MAX_TOKENS."] - S6["6. Resolve line numbers
from existing_code against diffs.
Re-locate via LLM if needed."] - S7["7. Emit text or JSON output
(and persist session to disk)"] - - Start --> S1 --> S2 --> S3 --> S4 --> S5 --> S6 --> S7 -``` - -## Project layout - -``` -open-code-review/ -├── cmd/opencodereview/ # CLI entry point: dispatch, flags, commands -├── internal/ -│ ├── agent/ # Per-file sub-agent loop + memory compression -│ ├── config/ -│ │ ├── allowlist/ # Default file-extension allowlist & exclusions -│ │ ├── rules/ # Layered rule resolver, system rule docs -│ │ ├── template/ # Plan / main / memory_compression prompts -│ │ ├── testconnection/ # Built-in `ocr llm test` task -│ │ └── toolsconfig/ # Tool definitions sent to the model -│ ├── diff/ # Git diff parsing, hunk math, relocation -│ ├── gitcmd/ # Git subprocess runner -│ ├── llm/ # Anthropic + OpenAI protocols, retries, BPE tokens -│ ├── model/ # Diff / Comment data structures -│ ├── pathutil/ # Path utilities -│ ├── release/ # Release-notes generation -│ ├── session/ # JSONL persistence of every review session -│ ├── stdout/ # Quiet-able stdout writer for `--audience agent` -│ ├── suggestdiff/ # Build "Apply suggestion" diffs -│ ├── telemetry/ # OpenTelemetry spans, metrics, exporters -│ ├── tool/ # The six built-in tools + comment collector -│ └── viewer/ # `ocr viewer` — local web UI for past sessions -├── pages/ # React-based marketing landing page (separate) -├── plugins/ # Claude Code plugin manifest + commands -├── extensions/ # Editor extensions (VS Code) -├── examples/ # CI recipes (GitHub Actions, GitLab CI) -├── skills/ # Generic agent Skill manifest -├── scripts/ # NPM install/update helpers, publish scripts -├── npm/ # Per-platform optional dependency packages -└── bin/ # NPM wrapper that shells out to the binary -``` - ## See Also - [QuickStart](../quickstart/) — install and run your first review. diff --git a/pages/src/content/docs/ja/overview.md b/pages/src/content/docs/ja/overview.md index 526149f..fcb602e 100644 --- a/pages/src/content/docs/ja/overview.md +++ b/pages/src/content/docs/ja/overview.md @@ -67,57 +67,6 @@ agent の強みは、最も重要な部分に集約されます。 （呼び出し頻度の分布、単一ツールの繰り返し率、各ツールが呼び出しチェーン全体に与える影響）。最終的に得られた 専用の [6 ツール](../tools/) セットは、汎用 agent のツールキットよりも安定していて予測可能です。 -## パイプラインの連携 - -```mermaid -flowchart TD - Start["ocr review --from main --to feature"] - S1["1. Resolve LLM endpoint
config / env / shell rc"] - S2["2. Load diffs from git
workspace / commit / range"] - S3["3. Filter files
binary → user_exclude → user_include
→ ext allowlist → default path"] - S4["4. Drop diffs > 80% of MAX_TOKENS"] - S5["5. Dispatch per-file sub-agents (concurrent)

For each file:
  a. Plan phase (if changed lines ≥ 50)
  b. Main loop: LLM → tool calls → … → task_done
  c. code_comment results collected (async via worker pool)

Memory compression triggers when context
exceeds 60 % (async) or 80 % (sync) of MAX_TOKENS."] - S6["6. Resolve line numbers
from existing_code against diffs.
Re-locate via LLM if needed."] - S7["7. Emit text or JSON output
(and persist session to disk)"] - - Start --> S1 --> S2 --> S3 --> S4 --> S5 --> S6 --> S7 -``` - -## プロジェクト構成 - -``` -open-code-review/ -├── cmd/opencodereview/ # CLI エントリ：ディスパッチ、引数、コマンド -├── internal/ -│ ├── agent/ # ファイルごとのサブ agent ループ + メモリ圧縮 -│ ├── config/ -│ │ ├── allowlist/ # デフォルトのファイル拡張子ホワイトリストと除外項目 -│ │ ├── rules/ # 階層型ルールパーサー、システムルールドキュメント -│ │ ├── template/ # plan / main / memory_compression prompt -│ │ ├── testconnection/ # 組み込みの `ocr llm test` タスク -│ │ └── toolsconfig/ # モデルに送信するツール定義 -│ ├── diff/ # Git diff 解析、hunk 計算、再配置 -│ ├── gitcmd/ # Git サブプロセスランナー -│ ├── llm/ # Anthropic + OpenAI プロトコル、リトライ、BPE token -│ ├── model/ # diff / コメント のデータ構造 -│ ├── pathutil/ # パスユーティリティ -│ ├── release/ # Release notes 生成 -│ ├── session/ # レビューセッションごとの JSONL 永続化 -│ ├── stdout/ # `--audience agent` 下でミュート可能な stdout writer -│ ├── suggestdiff/ # "Apply suggestion" diff の構築 -│ ├── telemetry/ # OpenTelemetry span、metrics、exporter -│ ├── tool/ # 6 つの組み込みツール + コメントコレクター -│ └── viewer/ # `ocr viewer`——過去のセッション用のローカル Web UI -├── pages/ # React ベースのマーケティング用ランディングページ（独立） -├── plugins/ # Claude Code プラグインマニフェスト + コマンド -├── extensions/ # エディタ拡張（VS Code） -├── examples/ # CI レシピ（GitHub Actions、GitLab CI） -├── skills/ # 汎用 agent Skill マニフェスト -├── scripts/ # NPM インストール/更新ヘルパー、リリーススクリプト -├── npm/ # 各プラットフォーム向け optional dependency パッケージ -└── bin/ # NPM wrapper、シェルからバイナリを呼び出す -``` - ## 関連項目 - [クイックスタート](../quickstart/)——インストールして初回のレビューを完了します。 diff --git a/pages/src/content/docs/zh/overview.md b/pages/src/content/docs/zh/overview.md index 3f036cb..f9f6df4 100644 --- a/pages/src/content/docs/zh/overview.md +++ b/pages/src/content/docs/zh/overview.md @@ -66,57 +66,6 @@ agent 的优势集中在最关键的地方： （调用频次分布、单工具重复率、每个工具对整体调用链的影响）。最终得到一套 专用 [六工具](../tools/) 集，比通用 agent 工具包更稳定、更可预测。 -## 流水线如何衔接 - -```mermaid -flowchart TD - Start["ocr review --from main --to feature"] - S1["1. Resolve LLM endpoint
config / env / shell rc"] - S2["2. Load diffs from git
workspace / commit / range"] - S3["3. Filter files
binary → user_exclude → user_include
→ ext allowlist → default path"] - S4["4. Drop diffs > 80% of MAX_TOKENS"] - S5["5. Dispatch per-file sub-agents (concurrent)

For each file:
  a. Plan phase (if changed lines ≥ 50)
  b. Main loop: LLM → tool calls → … → task_done
  c. code_comment results collected (async via worker pool)

Memory compression triggers when context
exceeds 60 % (async) or 80 % (sync) of MAX_TOKENS."] - S6["6. Resolve line numbers
from existing_code against diffs.
Re-locate via LLM if needed."] - S7["7. Emit text or JSON output
(and persist session to disk)"] - - Start --> S1 --> S2 --> S3 --> S4 --> S5 --> S6 --> S7 -``` - -## 项目结构 - -``` -open-code-review/ -├── cmd/opencodereview/ # CLI 入口：分发、参数、命令 -├── internal/ -│ ├── agent/ # 每文件子 agent 循环 + 记忆压缩 -│ ├── config/ -│ │ ├── allowlist/ # 默认文件扩展名白名单与排除项 -│ │ ├── rules/ # 分层规则解析器、系统规则文档 -│ │ ├── template/ # plan / main / memory_compression prompt -│ │ ├── testconnection/ # 内置 `ocr llm test` 任务 -│ │ └── toolsconfig/ # 发送给模型的工具定义 -│ ├── diff/ # Git diff 解析、hunk 数学、重新定位 -│ ├── gitcmd/ # Git 子进程运行器 -│ ├── llm/ # Anthropic + OpenAI 协议、重试、BPE token -│ ├── model/ # diff / 评论 数据结构 -│ ├── pathutil/ # 路径工具 -│ ├── release/ # Release notes 生成 -│ ├── session/ # 每次评审会话的 JSONL 持久化 -│ ├── stdout/ # `--audience agent` 下可静音的 stdout writer -│ ├── suggestdiff/ # 构建 "Apply suggestion" diff -│ ├── telemetry/ # OpenTelemetry span、metrics、exporter -│ ├── tool/ # 六个内置工具 + 评论收集器 -│ └── viewer/ # `ocr viewer`——历史会话的本地 Web UI -├── pages/ # 基于 React 的营销落地页（独立） -├── plugins/ # Claude Code 插件清单 + 命令 -├── extensions/ # 编辑器扩展（VS Code） -├── examples/ # CI 配方（GitHub Actions、GitLab CI） -├── skills/ # 通用 agent Skill 清单 -├── scripts/ # NPM 安装/更新助手、发布脚本 -├── npm/ # 各平台 optional dependency 包 -└── bin/ # NPM wrapper，shell 调用二进制 -``` - ## 另见 - [快速开始](../quickstart/)——安装并完成首次评审。