ContextStream MCP Server
公式セッションを超えたAIコーディングアシスタントのための永続メモリとセマンティック検索
ドキュメント
Docs · セクションを参照
01 · セットアップウィザード
1つのコマンドで完全設定。
1つのコマンドを実行するだけで、認証（ブラウザ/デバイスログイン）、APIキーの作成、ルールの生成、そしてお使いのツール（VS Codeの servers スキーマを含む）に適したMCP設定の書き込みが行われます。
terminal · terminal · macOS / Linux
curl -fsSL https://contextstream.io/scripts/mcp.sh | bash
terminal · powershell · Windows
irm https://contextstream.io/scripts/mcp.ps1 | iex
インストール済みですか？ セットアップウィザードを再実行
terminal · terminal · 再実行
contextstream-mcp setup
実行内容: MCP設定（VS Code servers、Cursor/Clineなど mcpServers）の書き込み、ルール（標準/拡張）の生成、プロジェクトのワークスペースへのリンクを行います。
統合ツールセット: ウィザードはデフォルトで約11個の統合ドメインツールを設定します（約75%のトークン削減）。オプション: router モード（約2つのメタツール、最もコンパクト）。全ツールカタログを参照してください。
ファイルを書き込まずに変更をプレビュー: contextstream-mcp setup --dry-run
チームワークフロー
共有メモリ、スキル、コンテキストの表示。
チームアカウントは共有プロジェクト以上のものを提供します。contextstream-mcp setup 中に、ウィザードはチーム機能を検出し、ワークスペースのヒントを表示します。エディタでは、すべての session(action="context") 呼び出しに、チームの推奨事項、ガバナンスの手がかり、優先度シグナル、リンクされたアーティファクトを含めることができます。
共有ワークスペースの選択
セットアップ中に各リポジトリをチームワークスペースにリンクすることで、インデックス、決定事項、チケットの整合性が保たれます。
チームスキルの共有
skill(action="share", scope="team") は、session(action="context") でチームメイトが自動的にマッチする再利用可能なワークフローを公開します。
実行スコープの切り替え
デュアルコンテキストアカウントは、MCPで --account-mode=team|personal|auto または session set_account_mode を使用します。
エンティティによる作業の追跡
チケット、ハンドオフ、インシデント、リリースは、インデックス付き参照を使用します — セッションやチームメイト間で永続します。
ホスト型リモートがデフォルトです。 ローカルバイナリMCPはリカバリ専用です — 明示的に必要な場合に CONTEXTSTREAM_ALLOW_LOCAL_MCP=1 を設定してください。チームセットアップ（招待/ロール）とログイン後チェックリストを参照してください。
CLIショートカット
非対話型コマンド（CIと更新）。
インタラクティブウィザードなしでこれらを実行します — アップグレード後、チームオンボーディング、認証情報のローテーション、ワークスペース変更時に最適です。contextstream-mcp --help でも表示されます。
| コマンド | 使用するタイミング |
|---|---|
| contextstream-mcp update-hooks --scope=global | アップグレード後またはチームワークスペース参加後 — PreToolUse/UserPromptSubmitフックを更新。 |
| contextstream-mcp update-rules --scope=all | 最新のチームワークフローガイダンスで .cursorrules / CLAUDE.md / AGENTS.md を再生成。 |
| contextstream-mcp update-configs --scope=global | APIキーまたはワークスペース変更後にMCP設定を再書き込み。 |
| contextstream-mcp migrate-remote --scope=all | レガシーなローカルstdio設定をホスト型リモートトランスポートに変換。 |
| contextstream-mcp detect-editors --format=json | インストールされているエディタをスクリプトで検出（ブートストラップ/CI）。 |
| contextstream-mcp generate-configs --transport=remote --preauth | ファイルを書き込まずにJSON設定ペイロードを出力。 |
| contextstream-mcp configure --transcripts=on --scope=all | トランスクリプトキャプチャのデフォルトを非対話的に設定。 |
terminal · アカウントモード · チーム vs 個人
# Default: follow account (auto)
contextstream-mcp --account-mode=auto
# Force team-scoped reads/writes
contextstream-mcp --account-mode=team
# Or set once in shell profile:
export CONTEXTSTREAM_ACCOUNT_MODE=team
02 · 手動設定
クライアントごとの設定。
クライアントごとに正しい形式を使用してください（VS Codeは servers を使用し、他の多くのクライアントは mcpServers を使用します）。
統合ツールセット: デフォルトでは、サーバーは約11個の統合ドメインツールを公開します（レガシーな細分化ツールと比較して約75%のトークン削減）。さらに少ないツールにするには、env ブロックに "CONTEXTSTREAM_PROGRESSIVE_MODE": "true" を追加して、約2つのルーターメタツールにします。全ツールカタログを参照してください。
お使いのツールにジャンプ
Cursor / VS CodeWindsurfCodex CLIOpenCode CLIClaude CodeClaude DesktopClineKilo CodeRoo CodeAntigravity
MCPとは？
AIのためのオープンプロトコル。
Model Context Protocol (MCP) は、AIアシスタントが外部ツールやデータソースに接続できるようにするオープンスタンダードです。ContextStreamのMCPサーバーを使用すると、AIツールは次のことが可能になります。
- 会話や決定事項をセッション間で記憶する
- コードベースやドキュメントを意味的に検索する
- 知識グラフを構築してクエリする
- 異なるAIツール間でコンテキストを共有する
自然言語
尋ねるだけ。AIがツールを処理します。
ツール名を覚えたり、直接呼び出したりする必要はありません。平易な英語でやりたいことを説明するだけで、AIアシスタントが自動的に適切なツールを使用します。
自然に尋ねるだけ
- · "セッションサマリー"
- · "認証について何を決めた？"
- · "PostgreSQLを使っていることを覚えておいて"
- · "支払いコードを検索して"
AIが残りを処理
- · 関連するコンテキストを自動的に検索
- · 必要に応じて過去の決定事項を呼び出す
- · 重要な情報をメモリに保存
- · コードとドキュメントを検索
例 · "セッションサマリー"
AIはあなたの意図を理解し、背後で適切なContextStreamツールを呼び出します。
前提条件
必要なもの。
- ContextStreamアカウント（セットアップウィザードでブラウザログインによりAPIキーを作成できます）。
コンテキストの強化
GitHub + Slack 統合
MCPはAIに永続的なメモリを提供します。GitHubとSlackを接続することで、そのメモリはさらに豊かになります — AIは質問に答える際に、PR、Issue、チームのディスカッションを自動的に参照できます。
自動コンテキスト強化
context_smart または session_smart_search を呼び出すと、関連するGitHubのIssue、PR、Slackのディスカッションが自動的に含まれます。追加のツールは不要です。
GitHubSync Issue、PR、リリース、コメント。ディスカッションから決定事項が自動的に抽出されます。SlackSync チャンネルとスレッド。エンゲージメントの高い会話がスコアリングされ、優先順位付けされます。
プロンプト例
- · "認証について何を決めた？" — GitHubのIssue + Slackのスレッドから決定事項を見つける
- · "支払いシステムに関する最近のアクティビティを見せて" — PR、Issue、チームディスカッションを表示
- · "前回の障害から何を学んだ？" — SlackとGitHubから洞察を取得
- · "GitHubアクティビティの週次サマリーをくれ" —
integration(provider="github", action="summary", ...)を使用 - · "データベース移行のディスカッションについてすべての統合を検索" —
integration(provider="all", action="search", ...)を使用 - · "すべてのソースにわたる週次チームサマリーをくれ" —
integration(provider="all", action="summary", ...)を使用
統合ツールアクションクイックリファレンス
integration(provider="github|slack|all", action="...") を使用
GitHub (provider="github")
action="stats"— 統計と同期ステータスaction="search"— 状態/期間フィルターで検索action="activity"— アクティビティフィード（日数フィルター）action="knowledge"— 抽出された決定事項/教訓action="summary"— 週次/月次サマリーaction="repos"— 同期されたリポジトリの一覧action="issues"— Issue/PRの一覧
Slack (provider="slack")
action="stats"— 統計と同期ステータスaction="search"— チャンネル/期間フィルターで検索action="discussions"— エンゲージメントの高いスレッドaction="knowledge"— 抽出された決定事項/教訓action="summary"— 週次/月次サマリーaction="channels"— 同期されたチャンネルの一覧
クロスソース (provider="all")
action="status"— 接続されたすべての統合の同期ステータスとヘルスをチェックaction="search"— 1つのクエリで接続されたすべての統合を検索action="summary"— すべてのソースにわたる統合アクティビティサマリー（日数フィルター）action="knowledge"— すべてのソースから決定事項、教訓、洞察を取得
クライアント · Cursor / VS Code
Cursor / VS Code
CursorとVS Codeは異なるMCP設定スキーマを使用します。Cursorは依然としてローカルMCPプロセスを使用しますが、VS Code/Copilotはホスト型ContextStream MCPをHTTP経由で直接使用できるようになりました。
terminal · .cursor/mcp.json (プロジェクト) または ~/.cursor/mcp.json (グローバル)
{
"mcpServers": {
"contextstream": {
"command": "contextstream-mcp",
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
VS Code / Copilotに推奨: ワンクリックリモートインストール
ホスト型ContextStream MCPをインストールし、初回使用時にVS CodeにOAuthを処理させます。ローカルバイナリやAPIキーを .vscode/mcp.json に保持する必要はありません。
VS CodeにインストールVS Code MCP ドキュメント
terminal · .vscode/mcp.json (VS CodeネイティブMCP, リモート)
{
"servers": {
"contextstream": {
"type": "http",
"url": "https://mcp.contextstream.io/mcp?default_context_mode=fast"
}
}
}
コマンドラインを使用しますか？ code --add-mcp でリモートサーバーを追加
terminal · terminal
code --add-mcp '{"name":"contextstream","type":"http","url":"https://mcp.contextstream.io/mcp?default_context_mode=fast"}'
初回使用時に、VS CodeはContextStreamの認証を求め、その後自動的にセットアップを完了するはずです。
セルフホストですか？ 同じリモート設定を https://mcp.contextstream.io/mcp?default_context_mode=fast の代わりに自身のMCPゲートウェイURLに向けます。
クライアント · OpenCode CLI
OpenCode CLI
ContextStreamをOpenCode CLIで使用するには、MCPサーバー設定を ~/.config/opencode/opencode.json ファイル（またはプロジェクトルートの opencode.json）に追加します。
terminal · ~/.config/opencode/opencode.json
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"contextstream": {
"type": "local",
"command": ["contextstream-mcp"],
"enabled": true,
"environment": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
設定を編集した後、OpenCodeを再起動してContextStream MCPサーバーをロードします。
クライアント · Codex CLI
Codex CLI
ContextStreamをCodex CLIで使用するには、MCPサーバー設定を ~/.codex/config.toml ファイルに追加します。
terminal · ~/.codex/config.toml
[mcp_servers.contextstream]
command = "contextstream-mcp"
args = []
[mcp_servers.contextstream.env]
CONTEXTSTREAM_API_URL = "https://api.contextstream.io"
CONTEXTSTREAM_API_KEY = "your_api_key"
設定を編集した後、Codexを再起動してContextStream MCPサーバーをロードします。
クライアント · Claude Code
Claude Code (CLI)
プロジェクトディレクトリからこのコマンドを実行して、ContextStreamをClaude Codeに追加します。
terminal · terminal
claude mcp add --transport stdio contextstream --env CONTEXTSTREAM_API_URL=https://api.contextstream.io --env CONTEXTSTREAM_API_KEY=your_api_key -- contextstream-mcp
Windowsの注意点（WSLではなくネイティブWindows）: -- の後に cmd /c contextstream-mcp を使用します。
代替手段: add-json (stdio)
terminal · terminal · add-json
claude mcp add-json contextstream \
'{"type":"stdio","command":"contextstream-mcp","args":[],"env":{"CONTEXTSTREAM_API_URL":"https://api.contextstream.io","CONTEXTSTREAM_API_KEY":"your_api_key"}}'
ヒント: チーム設定では、キーをシェル履歴に埋め込む代わりに、コミットされた .mcp.json ファイル（プロジェクトスコープ）を推奨します。
これにより、ContextStreamがプロジェクトパス下の ~/.claude.json に追加され、そのディレクトリで作業する際に利用可能になります。
MCPスコープオプション
- · ローカル (デフォルト): 自分専用、現在のプロジェクトのみ → プロジェクトパス下の
~/.claude.jsonに保存。 - · ユーザー (
--scope user): 自分専用、すべてのプロジェクト → グローバルに~/.claude.jsonに保存。 - · プロジェクト (
--scope project): チームと共有 → プロジェクトルートの.mcp.jsonに保存（gitにコミット）。
チーム共有には、プロジェクトスコープを使用して、gitにコミットできる .mcp.json ファイルを作成します。
terminal · .mcp.json (プロジェクトスコープ)
{
"mcpServers": {
"contextstream": {
"command": "contextstream-mcp",
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
MCPサーバーを追加した後、変更を有効にするためにClaude Codeを再起動します。claude mcp list でサーバーがロードされていることを確認します。.mcp.json からのプロジェクトスコープサーバーの場合、Claude Codeは初回使用時に承認を求めます。
クライアント · Claude Desktop
Claude Desktop (GUIアプリ)
ContextStreamをClaude Desktopアプリケーションに追加します。
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
terminal · claude_desktop_config.json
{
"mcpServers": {
"contextstream": {
"command": "contextstream-mcp",
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
設定を編集した後、Claude Desktopを終了して再起動し、変更を有効にします。
クライアント · Windsurf
Windsurf
グローバルMCP設定ファイルを編集して、ContextStreamをWindsurfに追加します。
設定: ~/.codeium/windsurf/mcp_config.json
terminal · ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"contextstream": {
"command": "contextstream-mcp",
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
ルールファイル
- · グローバル:
~/.codeium/windsurf/memories/global_rules.md - · プロジェクト:
.windsurf/rules/contextstream.md
Windsurfは、ContextStreamルールの自動適用のためのフックをサポートしています。設定を編集した後、Windsurfを再起動して変更を有効にします。
クライアント · Cline
Cline
ContextStreamをCline MCP設定に追加します。ClineでMCPサーバーアイコンをクリックし、「設定」タブを選択してから「MCPサーバーを設定」をクリックして編集します。
terminal · cline_mcp_settings.json
{
"mcpServers": {
"contextstream": {
"command": "contextstream-mcp",
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
設定を編集した後は、変更を反映させるためにClineを再起動してください。また、alwaysAllow を使用して特定のツールを自動承認することもできます。
クライアント · Kilo Code
Kilo Code
Kilo CodeのMCP設定にContextStreamを追加します。MCPサーバーはグローバルまたはプロジェクトごとに設定できます。
グローバル: Settings → MCP Servers → Installed → Edit Global MCP をクリックして mcp_settings.json を開きます
プロジェクト: プロジェクトルートの .kilocode/mcp.json
terminal · .kilocode/mcp.json (または mcp_settings.json)
{
"mcpServers": {
"contextstream": {
"command": "contextstream-mcp",
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
プロジェクトレベルの設定はグローバル設定よりも優先されます。編集後はKilo Codeを再起動してください。
クライアント · Roo Code
Roo Code
Roo CodeのMCP設定にContextStreamを追加します。MCPサーバーはグローバルまたはプロジェクトごとに設定できます。
グローバル: 設定アイコンをクリック → Edit Global MCP で mcp_settings.json を開きます
プロジェクト: プロジェクトルートの .roo/mcp.json
terminal · .roo/mcp.json (または mcp_settings.json)
{
"mcpServers": {
"contextstream": {
"command": "contextstream-mcp",
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
プロジェクトレベルの設定はグローバル設定よりも優先されます。編集後はRoo Codeを再起動してください。
クライアント · Antigravity
Antigravity (Google)
AntigravityはCursor/Claude Desktopと同じ形式のプロジェクトスコープの .mcp.json ファイルを使用します。
terminal · .mcp.json (プロジェクトルート)
{
"mcpServers": {
"contextstream": {
"command": "contextstream-mcp",
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
Windowsユーザー: cmdラッパーを使用
terminal · .mcp.json (Windows)
{
"mcpServers": {
"contextstream": {
"command": "cmd",
"args": ["/c", "contextstream-mcp"],
"env": {
"CONTEXTSTREAM_API_URL": "https://api.contextstream.io",
"CONTEXTSTREAM_API_KEY": "your_api_key"
}
}
}
}
ルールファイル
- · グローバル:
~/.gemini/GEMINI.md - · ワークスペース:
.agent/rules/contextstream.md
「...」メニュー → 「MCP Servers」 → 「View raw config」から設定を確認できます。編集後はAntigravityを再起動してください。
エディタールール
自動的なContextStream利用の改善
推奨
ContextStreamの自動コンテキスト機能は、ContextStreamツールを使用するたびにワークスペースのコンテキストを自動的に読み込みます。しかし、AIアシスタントが常に積極的に決定を保存したり、過去のコンテキストを思い出したりするとは限りません。エディターAIルールを追加することで一貫性が向上し、会話全体を通じてAIが決定、好み、重要なコンテキストを自動的にキャプチャするようになります。
重要: MCPツールの命名規則
AIツールによってMCPツールの命名規則が異なります。誤った形式を使用するとツールが見つからなくなります。
| AIツール | 形式 | 例 |
|---|---|---|
| Claude Code | mcp____ | mcp__contextstream__session_init |
| Codex CLI / OpenCode CLI | (生の名前) | session_init |
| Cursor / Windsurf / Cline | (生の名前) | session_init |
| Kilo Code / Roo Code | (生の名前) | session_init |
要約: Claude Codeのみが mcp__contextstream__ プレフィックスを使用します。他のすべてのツールは生のツール名を使用します。
ContextStreamルールは2つのレベルで追加できます: グローバル (すべてのプロジェクトに適用) または プロジェクト (1つのプロジェクトに適用)。
グローバルルール (全プロジェクト)
これらを一度追加すれば、すべてのプロジェクトに自動的に適用されます。
| エディター | グローバルルールの場所 |
|---|---|
| Cursor | Settings → General → Rules for AI |
| Windsurf | ~/.codeium/windsurf/memories/global_rules.md |
| Cline | ~/Documents/Cline/Rules/ |
| Kilo Code | ~/.kilocode/rules/ |
| Roo Code | ~/.roo/rules/ |
| Claude Code | ~/.claude/CLAUDE.md |
| Codex CLI | ~/.codex/AGENTS.md (グローバル) または親フォルダ (例: ~/dev/AGENTS.md) |
| OpenCode CLI | ~/.config/opencode/AGENTS.md |
プロジェクトルール (単一プロジェクト)
特定のプロジェクトに追加します。Cursorのフォルダベースのルールでは、ルールが常にアクティブになるようにアクティベーションモードを "Always On" に設定してください。
| エディター | プロジェクトルールの場所 |
|---|---|
| Cursor | .cursorrules または .cursor/rules/*.mdc |
| Windsurf | .windsurf/rules/contextstream.md |
| Cline | .clinerules ファイルまたは .clinerules/ フォルダ |
| Kilo Code | .kilocode/rules/ |
| Roo Code | .roo/rules/contextstream.md または .roo/rules/ フォルダ |
| Claude Code | プロジェクトルートの CLAUDE.md |
| Codex CLI | プロジェクトルートの AGENTS.md |
| OpenCode CLI | プロジェクトルートの AGENTS.md |
| Aider | プロジェクトルートの .aider.conf.yml |
アクティベーションモード (Cursor, Kilo Code & Roo Code)
フォルダベースのルール (例: .cursor/rules/) を使用する場合、各ルールファイルにはアクティベーションモードがあります。
- Always On — 常にアクティブ (ContextStreamに推奨)
- Manual — ルールを@メンションしたときのみ
- Model Decision — AIが説明に基づいて判断
- Glob — 一致するファイルパターンでアクティブ
グローバルルールとルートレベルのファイル (.cursorrules) は常にアクティブです。
セットアップウィザードを使用しますか？ 最初に実行してください。それ以外の場合は、これらの標準ルールを手動で追加してください。各エディターの例:
Claude Code
プロジェクトルートに CLAUDE.md ファイルを作成するか、グローバルの ~/.claude/CLAUDE.md に追加します。
terminal · CLAUDE.md (標準)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `mcp__contextstream__search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `mcp__contextstream__context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x (Hooks Enforced)
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
### Required Every Message
| Action | Tool Call |
|--------|-----------|
| **1st message** | `mcp__contextstream__session_init(folder_path="<cwd>", context_hint="<msg>")` then `mcp__contextstream__context_smart(...)` |
| **2nd+ messages** | `mcp__contextstream__context_smart(user_message="<msg>", format="minified", max_tokens=400)` |
| **Code search** | `mcp__contextstream__search(mode="hybrid", query="...")` — BEFORE any local tools |
| **Save decisions** | `mcp__contextstream__session(action="capture", event_type="decision", ...)` |
### Search Modes
| Mode | Use Case |
|------|----------|
| `hybrid` | General code mcp__contextstream__search (default) |
| `keyword` | Exact symbol/string match |
| `exhaustive` | Find ALL matches (grep-like) |
| `semantic` | Conceptual questions |
### Why ContextStream First?
❌ **WRONG:** `Grep → Read → Read → Read` (4+ tool calls, slow)
✅ **CORRECT:** `mcp__contextstream__search(mode="hybrid")` (1 call, returns context)
ContextStream search is **indexed** and returns semantic matches + context in ONE call.
### Quick Reference
| Tool | Example |
|------|---------|
| `search` | `mcp__contextstream__search(mode="hybrid", query="auth", limit=3)` |
| `session` | `mcp__contextstream__session(action="capture", event_type="decision", title="...", content="...")` |
| `memory` | `mcp__contextstream__memory(action="list_events", limit=10)` |
| `graph` | `mcp__contextstream__graph(action="dependencies", file_path="...")` |
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `mcp__contextstream__session(action="get_lessons", query="<topic>")`
- On mistakes: `mcp__contextstream__session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
### Plans & Tasks
When user asks for a plan, use ContextStream (not EnterPlanMode):
1. `mcp__contextstream__session(action="capture_plan", title="...", steps=[...])`
2. `mcp__contextstream__memory(action="create_task", title="...", plan_id="<id>")`
Full docs: https://contextstream.io/docs/mcp/tools
拡張ルールを表示 (詳細)
terminal · CLAUDE.md (拡張)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `mcp__contextstream__search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `mcp__contextstream__context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x Integration (Enhanced)
You have access to ContextStream MCP tools for persistent memory and context.
v0.4.x uses **~11 consolidated domain tools** for ~75% token reduction vs previous versions.
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
## TL;DR - REQUIRED EVERY MESSAGE
| Message | What to Call |
|---------|--------------|
| **1st message** | `mcp__contextstream__session_init(folder_path="...", context_hint="<user's message>")`, then `mcp__contextstream__context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **2nd+ messages** | `mcp__contextstream__context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **🔍 ANY code search** | `mcp__contextstream__search(mode="hybrid", query="...")` — ALWAYS before Glob/Grep/Search/Read |
| **Before risky/non-trivial work** | `mcp__contextstream__session(action="get_lessons", query="<topic>")` |
| **After completing task** | `mcp__contextstream__session(action="capture", event_type="decision", ...)` - MUST capture |
| **User frustration/correction** | `mcp__contextstream__session(action="capture_lesson", ...)` - MUST capture lessons |
**NO EXCEPTIONS.** Do not skip even if you think you have enough context.
**First message rule:** After `session_init`, always call `context_smart` before any other tool or response.
**Context Pack (Pro+):** If enabled, use `mcp__contextstream__context_smart(..., mode="pack", distill=true)` for code/file queries. If unavailable or disabled, omit `mode` and proceed with standard `context_smart` (the API will fall back).
**Tool naming:** Use the exact tool names exposed by your MCP client. Claude Code typically uses `mcp__<server>__<tool>` where `<server>` matches your MCP config (often `contextstream`). If a tool call fails with "No such tool available", refresh rules and match the tool list.
---
## Consolidated Domain Tools Architecture
v0.4.x consolidates ~58 individual tools into ~11 domain tools with action/mode dispatch:
### Standalone Tools (Always Call)
- **`session_init`** - Initialize session with workspace detection + context
- **`context_smart`** - Semantic search for relevant context (CALL EVERY MESSAGE, including immediately after `session_init`)
### Domain Tools (Use action/mode parameter)
| Domain | Actions/Modes | Example |
|--------|---------------|---------|
| **`search`** | mode: semantic, hybrid, keyword, pattern, exhaustive, refactor | `mcp__contextstream__search(mode="hybrid", query="auth implementation", limit=3)` |
| **`session`** | action: capture, capture_lesson, get_lessons, recall, remember, user_context, summary, compress, delta, smart_search, decision_trace | `mcp__contextstream__session(action="capture", event_type="decision", title="Use JWT", content="...")` |
| **`memory`** | action: create_event, get_event, update_event, delete_event, list_events, distill_event, create_node, get_node, update_node, delete_node, list_nodes, supersede_node, search, decisions, timeline, summary | `mcp__contextstream__memory(action="list_events", limit=10)` |
| **`graph`** | action: dependencies, impact, call_path, related, path, decisions, ingest, circular_dependencies, unused_code, contradictions | `mcp__contextstream__graph(action="impact", symbol_name="AuthService")` |
| **`project`** | action: list, get, create, update, index, overview, statistics, files, index_status, ingest_local | `mcp__contextstream__project(action="statistics")` |
| **`workspace`** | action: list, get, associate, bootstrap | `mcp__contextstream__workspace(action="list")` |
| **`reminder`** | action: list, active, create, snooze, complete, dismiss | `mcp__contextstream__reminder(action="active")` |
| **`integration`** | provider: slack/github/all; action: status, search, stats, activity, contributors, knowledge, summary, channels, discussions, sync_users, repos, issues | `mcp__contextstream__integration(provider="github", action="search", query="...")` |
| **`help`** | action: tools, auth, version, editor_rules, enable_bundle | `mcp__contextstream__help(action="tools")` |
---
### Why context_smart is Required (Even After session_init)
**Common mistake:** "session_init already gave me context, I don't need context_smart"
**This is WRONG. Here's why:**
- `session_init` returns the last ~10 items **BY TIME** (chronological)
- `context_smart` **SEARCHES** for items **RELEVANT to THIS message** (semantic)
**Example failure:**
- User asks: "how should I implement authentication?"
- Auth decisions were made 20 conversations ago
- `session_init` won't have it (too old, not in recent 10)
- `context_smart` FINDS it via semantic search
**Without context_smart, you WILL miss relevant older context.**
---
### Search & Code Intelligence (ContextStream-first)
⚠️ **STOP: Before using Search/Glob/Grep/Read/Explore** → Call `mcp__contextstream__search(mode="hybrid")` FIRST. Use local tools ONLY if ContextStream returns 0 results.
**❌ WRONG workflow (wastes tokens, slow):**
Grep "function" → Read file1.ts → Read file2.ts → Read file3.ts → ようやく理解
**✅ CORRECT workflow (fast, complete):**
mcp__contextstream__search(mode="hybrid", query="function implementation") → 完了 (結果にコンテキストが含まれる)
**Why?** ContextStream search returns semantic matches + context + file locations in ONE call. Local tools require multiple round-trips.
**Search Mode Selection:**
| Need | Mode | Example |
|------|------|---------|
| Find code by meaning | `hybrid` | "authentication logic", "error handling" |
| Exact string/symbol | `keyword` | "UserAuthService", "API_KEY" |
| File patterns | `pattern` | "*.sql", "test_*.py" |
| ALL matches (grep-like) | `exhaustive` | "TODO", "FIXME" (find all occurrences) |
| Symbol renaming | `refactor` | "oldFunctionName" (word-boundary matching) |
| Conceptual search | `semantic` | "how does caching work" |
**Token Efficiency:** Use `output_format` to reduce response size:
- `full` (default): Full content for understanding code
- `paths`: File paths only (80% token savings) - use for file listings
- `minimal`: Compact format (60% savings) - use for refactoring
- `count`: Match counts only (90% savings) - use for quick checks
---
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `mcp__contextstream__session(action="get_lessons", query="<topic>")`
- On mistakes: `mcp__contextstream__session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
---
### Plans & Tasks
When user asks to create a plan or implementation roadmap:
1. Create plan: `mcp__contextstream__session(action="capture_plan", title="Plan Title", description="...", goals=["goal1", "goal2"], steps=[{id: "1", title: "Step 1", order: 1}, ...])`
2. Get plan_id from response, then create tasks: `mcp__contextstream__memory(action="create_task", title="Task Title", plan_id="<plan_id>", priority="high|medium|low", description="...")`
To manage existing plans/tasks:
- List plans: `mcp__contextstream__session(action="list_plans")`
- Get plan with tasks: `mcp__contextstream__session(action="get_plan", plan_id="<uuid>", include_tasks=true)`
- List tasks: `mcp__contextstream__memory(action="list_tasks", plan_id="<uuid>")` or `mcp__contextstream__memory(action="list_tasks")` for all
- Update task status: `mcp__contextstream__memory(action="update_task", task_id="<uuid>", task_status="pending|in_progress|completed|blocked")`
---
### Rules Update Notices
- If you see **[RULES_NOTICE]**, update rules via `mcp__contextstream__generate_rules(overwrite_existing=true)` (preserves custom rules).
- If you see **[VERSION_NOTICE]**, tell the user to update MCP using the provided command.
See full documentation: https://contextstream.io/docs/mcp/tools
Codex CLI / OpenCode CLI
プロジェクトルート (プロジェクトルール) または ~/.codex/AGENTS.md (Codexグローバル) / ~/.config/opencode/AGENTS.md (OpenCodeグローバル) に AGENTS.md ファイルを作成します。
Codex / OpenCode vs Claude ツール名
Codex / OpenCode は生のMCPツール名 (例: session_init) を使用します。Claude Code は名前空間付きツール名 (例: mcp__contextstream__session_init) を使用します。AIツールに適した形式を使用してください。
terminal · AGENTS.md (標準)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x (Hooks Enforced)
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
### Required Every Message
| Action | Tool Call |
|--------|-----------|
| **1st message** | `session_init(folder_path="<cwd>", context_hint="<msg>")` then `context_smart(...)` |
| **2nd+ messages** | `context_smart(user_message="<msg>", format="minified", max_tokens=400)` |
| **Code search** | `search(mode="hybrid", query="...")` — BEFORE any local tools |
| **Save decisions** | `session(action="capture", event_type="decision", ...)` |
### Search Modes
| Mode | Use Case |
|------|----------|
| `hybrid` | General code search (default) |
| `keyword` | Exact symbol/string match |
| `exhaustive` | Find ALL matches (grep-like) |
| `semantic` | Conceptual questions |
### Why ContextStream First?
❌ **WRONG:** `Grep → Read → Read → Read` (4+ tool calls, slow)
✅ **CORRECT:** `search(mode="hybrid")` (1 call, returns context)
ContextStream search is **indexed** and returns semantic matches + context in ONE call.
### Quick Reference
| Tool | Example |
|------|---------|
| `search` | `search(mode="hybrid", query="auth", limit=3)` |
| `session` | `session(action="capture", event_type="decision", title="...", content="...")` |
| `memory` | `memory(action="list_events", limit=10)` |
| `graph` | `graph(action="dependencies", file_path="...")` |
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
### Plans & Tasks
When user asks for a plan, use ContextStream (not EnterPlanMode):
1. `session(action="capture_plan", title="...", steps=[...])`
2. `memory(action="create_task", title="...", plan_id="<id>")`
Full docs: https://contextstream.io/docs/mcp/tools
拡張ルールを表示 (詳細)
terminal · AGENTS.md (拡張)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x Integration (Enhanced)
You have access to ContextStream MCP tools for persistent memory and context.
v0.4.x uses **~11 consolidated domain tools** for ~75% token reduction vs previous versions.
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
## TL;DR - REQUIRED EVERY MESSAGE
| Message | What to Call |
|---------|--------------|
| **1st message** | `session_init(folder_path="...", context_hint="<user's message>")`, then `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **2nd+ messages** | `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **🔍 ANY code search** | `search(mode="hybrid", query="...")` — ALWAYS before Glob/Grep/Search/Read |
| **Before risky/non-trivial work** | `session(action="get_lessons", query="<topic>")` |
| **After completing task** | `session(action="capture", event_type="decision", ...)` - MUST capture |
| **User frustration/correction** | `session(action="capture_lesson", ...)` - MUST capture lessons |
**NO EXCEPTIONS.** Do not skip even if you think you have enough context.
**First message rule:** After `session_init`, always call `context_smart` before any other tool or response.
**Context Pack (Pro+):** If enabled, use `context_smart(..., mode="pack", distill=true)` for code/file queries. If unavailable or disabled, omit `mode` and proceed with standard `context_smart` (the API will fall back).
**Tool naming:** Use the exact tool names exposed by your MCP client. Claude Code typically uses `mcp__<server>__<tool>` where `<server>` matches your MCP config (often `contextstream`). If a tool call fails with "No such tool available", refresh rules and match the tool list.
---
## Consolidated Domain Tools Architecture
v0.4.x consolidates ~58 individual tools into ~11 domain tools with action/mode dispatch:
### Standalone Tools (Always Call)
- **`session_init`** - Initialize session with workspace detection + context
- **`context_smart`** - Semantic search for relevant context (CALL EVERY MESSAGE, including immediately after `session_init`)
### Domain Tools (Use action/mode parameter)
| Domain | Actions/Modes | Example |
|--------|---------------|---------|
| **`search`** | mode: semantic, hybrid, keyword, pattern, exhaustive, refactor | `search(mode="hybrid", query="auth implementation", limit=3)` |
| **`session`** | action: capture, capture_lesson, get_lessons, recall, remember, user_context, summary, compress, delta, smart_search, decision_trace | `session(action="capture", event_type="decision", title="Use JWT", content="...")` |
| **`memory`** | action: create_event, get_event, update_event, delete_event, list_events, distill_event, create_node, get_node, update_node, delete_node, list_nodes, supersede_node, search, decisions, timeline, summary | `memory(action="list_events", limit=10)` |
| **`graph`** | action: dependencies, impact, call_path, related, path, decisions, ingest, circular_dependencies, unused_code, contradictions | `graph(action="impact", symbol_name="AuthService")` |
| **`project`** | action: list, get, create, update, index, overview, statistics, files, index_status, ingest_local | `project(action="statistics")` |
| **`workspace`** | action: list, get, associate, bootstrap | `workspace(action="list")` |
| **`reminder`** | action: list, active, create, snooze, complete, dismiss | `reminder(action="active")` |
| **`integration`** | provider: slack/github/all; action: status, search, stats, activity, contributors, knowledge, summary, channels, discussions, sync_users, repos, issues | `integration(provider="github", action="search", query="...")` |
| **`help`** | action: tools, auth, version, editor_rules, enable_bundle | `help(action="tools")` |
---
### Why context_smart is Required (Even After session_init)
**Common mistake:** "session_init already gave me context, I don't need context_smart"
**This is WRONG. Here's why:**
- `session_init` returns the last ~10 items **BY TIME** (chronological)
- `context_smart` **SEARCHES** for items **RELEVANT to THIS message** (semantic)
**Example failure:**
- User asks: "how should I implement authentication?"
- Auth decisions were made 20 conversations ago
- `session_init` won't have it (too old, not in recent 10)
- `context_smart` FINDS it via semantic search
**Without context_smart, you WILL miss relevant older context.**
---
### Search & Code Intelligence (ContextStream-first)
⚠️ **STOP: Before using Search/Glob/Grep/Read/Explore** → Call `search(mode="hybrid")` FIRST. Use local tools ONLY if ContextStream returns 0 results.
**❌ WRONG workflow (wastes tokens, slow):**
Grep "function" → Read file1.ts → Read file2.ts → Read file3.ts → ようやく理解
**✅ CORRECT workflow (fast, complete):**
search(mode="hybrid", query="function implementation") → 完了 (結果にコンテキストが含まれる)
**Why?** ContextStream search returns semantic matches + context + file locations in ONE call. Local tools require multiple round-trips.
**Search Mode Selection:**
| Need | Mode | Example |
|------|------|---------|
| Find code by meaning | `hybrid` | "authentication logic", "error handling" |
| Exact string/symbol | `keyword` | "UserAuthService", "API_KEY" |
| File patterns | `pattern` | "*.sql", "test_*.py" |
| ALL matches (grep-like) | `exhaustive` | "TODO", "FIXME" (find all occurrences) |
| Symbol renaming | `refactor` | "oldFunctionName" (word-boundary matching) |
| Conceptual search | `semantic` | "how does caching work" |
**Token Efficiency:** Use `output_format` to reduce response size:
- `full` (default): Full content for understanding code
- `paths`: File paths only (80% token savings) - use for file listings
- `minimal`: Compact format (60% savings) - use for refactoring
- `count`: Match counts only (90% savings) - use for quick checks
---
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
---
### Plans & Tasks
When user asks to create a plan or implementation roadmap:
1. Create plan: `session(action="capture_plan", title="Plan Title", description="...", goals=["goal1", "goal2"], steps=[{id: "1", title: "Step 1", order: 1}, ...])`
2. Get plan_id from response, then create tasks: `memory(action="create_task", title="Task Title", plan_id="<plan_id>", priority="high|medium|low", description="...")`
To manage existing plans/tasks:
- List plans: `session(action="list_plans")`
- Get plan with tasks: `session(action="get_plan", plan_id="<uuid>", include_tasks=true)`
- List tasks: `memory(action="list_tasks", plan_id="<uuid>")` or `memory(action="list_tasks")` for all
- Update task status: `memory(action="update_task", task_id="<uuid>", task_status="pending|in_progress|completed|blocked")`
---
### Rules Update Notices
- If you see **[RULES_NOTICE]**, update rules via `generate_rules(overwrite_existing=true)` (preserves custom rules).
- If you see **[VERSION_NOTICE]**, tell the user to update MCP using the provided command.
See full documentation: https://contextstream.io/docs/mcp/tools
Windsurf
プロジェクトルートに .windsurf/rules/contextstream.md ファイルを作成するか、グローバルの ~/.codeium/windsurf/memories/global_rules.md に追加します。
terminal · .windsurf/rules/contextstream.md (標準)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x (Hooks Enforced)
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
### Required Every Message
| Action | Tool Call |
|--------|-----------|
| **1st message** | `session_init(folder_path="<cwd>", context_hint="<msg>")` then `context_smart(...)` |
| **2nd+ messages** | `context_smart(user_message="<msg>", format="minified", max_tokens=400)` |
| **Code search** | `search(mode="hybrid", query="...")` — BEFORE any local tools |
| **Save decisions** | `session(action="capture", event_type="decision", ...)` |
### Search Modes
| Mode | Use Case |
|------|----------|
| `hybrid` | General code search (default) |
| `keyword` | Exact symbol/string match |
| `exhaustive` | Find ALL matches (grep-like) |
| `semantic` | Conceptual questions |
### Why ContextStream First?
❌ **WRONG:** `Grep → Read → Read → Read` (4+ tool calls, slow)
✅ **CORRECT:** `search(mode="hybrid")` (1 call, returns context)
ContextStream search is **indexed** and returns semantic matches + context in ONE call.
### Quick Reference
| Tool | Example |
|------|---------|
| `search` | `search(mode="hybrid", query="auth", limit=3)` |
| `session` | `session(action="capture", event_type="decision", title="...", content="...")` |
| `memory` | `memory(action="list_events", limit=10)` |
| `graph` | `graph(action="dependencies", file_path="...")` |
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
### Plans & Tasks
When user asks for a plan, use ContextStream (not EnterPlanMode):
1. `session(action="capture_plan", title="...", steps=[...])`
2. `memory(action="create_task", title="...", plan_id="<id>")`
Full docs: https://contextstream.io/docs/mcp/tools
拡張ルールを表示 (詳細)
terminal · .windsurf/rules/contextstream.md (拡張)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x Integration (Enhanced)
You have access to ContextStream MCP tools for persistent memory and context.
v0.4.x uses **~11 consolidated domain tools** for ~75% token reduction vs previous versions.
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
## TL;DR - REQUIRED EVERY MESSAGE
| Message | What to Call |
|---------|--------------|
| **1st message** | `session_init(folder_path="...", context_hint="<user's message>")`, then `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **2nd+ messages** | `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **🔍 ANY code search** | `search(mode="hybrid", query="...")` — ALWAYS before Glob/Grep/Search/Read |
| **Before risky/non-trivial work** | `session(action="get_lessons", query="<topic>")` |
| **After completing task** | `session(action="capture", event_type="decision", ...)` - MUST capture |
| **User frustration/correction** | `session(action="capture_lesson", ...)` - MUST capture lessons |
**NO EXCEPTIONS.** Do not skip even if you think you have enough context.
**First message rule:** After `session_init`, always call `context_smart` before any other tool or response.
**Context Pack (Pro+):** If enabled, use `context_smart(..., mode="pack", distill=true)` for code/file queries. If unavailable or disabled, omit `mode` and proceed with standard `context_smart` (the API will fall back).
**Tool naming:** Use the exact tool names exposed by your MCP client. Claude Code typically uses `mcp__<server>__<tool>` where `<server>` matches your MCP config (often `contextstream`). If a tool call fails with "No such tool available", refresh rules and match the tool list.
---
## Consolidated Domain Tools Architecture
v0.4.x consolidates ~58 individual tools into ~11 domain tools with action/mode dispatch:
### Standalone Tools (Always Call)
- **`session_init`** - Initialize session with workspace detection + context
- **`context_smart`** - Semantic search for relevant context (CALL EVERY MESSAGE, including immediately after `session_init`)
### Domain Tools (Use action/mode parameter)
| Domain | Actions/Modes | Example |
|--------|---------------|---------|
| **`search`** | mode: semantic, hybrid, keyword, pattern, exhaustive, refactor | `search(mode="hybrid", query="auth implementation", limit=3)` |
| **`session`** | action: capture, capture_lesson, get_lessons, recall, remember, user_context, summary, compress, delta, smart_search, decision_trace | `session(action="capture", event_type="decision", title="Use JWT", content="...")` |
| **`memory`** | action: create_event, get_event, update_event, delete_event, list_events, distill_event, create_node, get_node, update_node, delete_node, list_nodes, supersede_node, search, decisions, timeline, summary | `memory(action="list_events", limit=10)` |
| **`graph`** | action: dependencies, impact, call_path, related, path, decisions, ingest, circular_dependencies, unused_code, contradictions | `graph(action="impact", symbol_name="AuthService")` |
| **`project`** | action: list, get, create, update, index, overview, statistics, files, index_status, ingest_local | `project(action="statistics")` |
| **`workspace`** | action: list, get, associate, bootstrap | `workspace(action="list")` |
| **`reminder`** | action: list, active, create, snooze, complete, dismiss | `reminder(action="active")` |
| **`integration`** | provider: slack/github/all; action: status, search, stats, activity, contributors, knowledge, summary, channels, discussions, sync_users, repos, issues | `integration(provider="github", action="search", query="...")` |
| **`help`** | action: tools, auth, version, editor_rules, enable_bundle | `help(action="tools")` |
---
### Why context_smart is Required (Even After session_init)
**Common mistake:** "session_init already gave me context, I don't need context_smart"
**This is WRONG. Here's why:**
- `session_init` returns the last ~10 items **BY TIME** (chronological)
- `context_smart` **SEARCHES** for items **RELEVANT to THIS message** (semantic)
**Example failure:**
- User asks: "how should I implement authentication?"
- Auth decisions were made 20 conversations ago
- `session_init` won't have it (too old, not in recent 10)
- `context_smart` FINDS it via semantic search
**Without context_smart, you WILL miss relevant older context.**
---
### Search & Code Intelligence (ContextStream-first)
⚠️ **STOP: Before using Search/Glob/Grep/Read/Explore** → Call `search(mode="hybrid")` FIRST. Use local tools ONLY if ContextStream returns 0 results.
**❌ WRONG workflow (wastes tokens, slow):**
Grep "function" → Read file1.ts → Read file2.ts → Read file3.ts → ようやく理解
**✅ CORRECT workflow (fast, complete):**
search(mode="hybrid", query="function implementation") → 完了 (結果にコンテキストが含まれる)
**Why?** ContextStream search returns semantic matches + context + file locations in ONE call. Local tools require multiple round-trips.
**Search Mode Selection:**
| Need | Mode | Example |
|------|------|---------|
| Find code by meaning | `hybrid` | "authentication logic", "error handling" |
| Exact string/symbol | `keyword` | "UserAuthService", "API_KEY" |
| File patterns | `pattern` | "*.sql", "test_*.py" |
| ALL matches (grep-like) | `exhaustive` | "TODO", "FIXME" (find all occurrences) |
| Symbol renaming | `refactor` | "oldFunctionName" (word-boundary matching) |
| Conceptual search | `semantic` | "how does caching work" |
**Token Efficiency:** Use `output_format` to reduce response size:
- `full` (default): Full content for understanding code
- `paths`: File paths only (80% token savings) - use for file listings
- `minimal`: Compact format (60% savings) - use for refactoring
- `count`: Match counts only (90% savings) - use for quick checks
---
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
---
### Plans & Tasks
When user asks to create a plan or implementation roadmap:
1. Create plan: `session(action="capture_plan", title="Plan Title", description="...", goals=["goal1", "goal2"], steps=[{id: "1", title: "Step 1", order: 1}, ...])`
2. Get plan_id from response, then create tasks: `memory(action="create_task", title="Task Title", plan_id="<plan_id>", priority="high|medium|low", description="...")`
To manage existing plans/tasks:
- List plans: `session(action="list_plans")`
- Get plan with tasks: `session(action="get_plan", plan_id="<uuid>", include_tasks=true)`
- List tasks: `memory(action="list_tasks", plan_id="<uuid>")` or `memory(action="list_tasks")` for all
- Update task status: `memory(action="update_task", task_id="<uuid>", task_status="pending|in_progress|completed|blocked")`
---
### Rules Update Notices
- If you see **[RULES_NOTICE]**, update rules via `generate_rules(overwrite_existing=true)` (preserves custom rules).
- If you see **[VERSION_NOTICE]**, tell the user to update MCP using the provided command.
See full documentation: https://contextstream.io/docs/mcp/tools
Kilo Code
.kilocode/rules/ にMarkdownファイルを作成します。
terminal · .kilocode/rules/contextstream.md (標準)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x (Hooks Enforced)
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
### Required Every Message
| Action | Tool Call |
|--------|-----------|
| **1st message** | `session_init(folder_path="<cwd>", context_hint="<msg>")` then `context_smart(...)` |
| **2nd+ messages** | `context_smart(user_message="<msg>", format="minified", max_tokens=400)` |
| **Code search** | `search(mode="hybrid", query="...")` — BEFORE any local tools |
| **Save decisions** | `session(action="capture", event_type="decision", ...)` |
### Search Modes
| Mode | Use Case |
|------|----------|
| `hybrid` | General code search (default) |
| `keyword` | Exact symbol/string match |
| `exhaustive` | Find ALL matches (grep-like) |
| `semantic` | Conceptual questions |
### Why ContextStream First?
❌ **WRONG:** `Grep → Read → Read → Read` (4+ tool calls, slow)
✅ **CORRECT:** `search(mode="hybrid")` (1 call, returns context)
ContextStream search is **indexed** and returns semantic matches + context in ONE call.
### Quick Reference
| Tool | Example |
|------|---------|
| `search` | `search(mode="hybrid", query="auth", limit=3)` |
| `session` | `session(action="capture", event_type="decision", title="...", content="...")` |
| `memory` | `memory(action="list_events", limit=10)` |
| `graph` | `graph(action="dependencies", file_path="...")` |
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
### Plans & Tasks
When user asks for a plan, use ContextStream (not EnterPlanMode):
1. `session(action="capture_plan", title="...", steps=[...])`
2. `memory(action="create_task", title="...", plan_id="<id>")`
Full docs: https://contextstream.io/docs/mcp/tools
拡張ルールを表示 (詳細)
terminal · .kilocode/rules/contextstream.md (拡張)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x Integration (Enhanced)
You have access to ContextStream MCP tools for persistent memory and context.
v0.4.x uses **~11 consolidated domain tools** for ~75% token reduction vs previous versions.
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
## TL;DR - REQUIRED EVERY MESSAGE
| Message | What to Call |
|---------|--------------|
| **1st message** | `session_init(folder_path="...", context_hint="<user's message>")`, then `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **2nd+ messages** | `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **🔍 ANY code search** | `search(mode="hybrid", query="...")` — ALWAYS before Glob/Grep/Search/Read |
| **Before risky/non-trivial work** | `session(action="get_lessons", query="<topic>")` |
| **After completing task** | `session(action="capture", event_type="decision", ...)` - MUST capture |
| **User frustration/correction** | `session(action="capture_lesson", ...)` - MUST capture lessons |
**NO EXCEPTIONS.** Do not skip even if you think you have enough context.
**First message rule:** After `session_init`, always call `context_smart` before any other tool or response.
**Context Pack (Pro+):** If enabled, use `context_smart(..., mode="pack", distill=true)` for code/file queries. If unavailable or disabled, omit `mode` and proceed with standard `context_smart` (the API will fall back).
**Tool naming:** Use the exact tool names exposed by your MCP client. Claude Code typically uses `mcp__<server>__<tool>` where `<server>` matches your MCP config (often `contextstream`). If a tool call fails with "No such tool available", refresh rules and match the tool list.
---
## Consolidated Domain Tools Architecture
v0.4.x consolidates ~58 individual tools into ~11 domain tools with action/mode dispatch:
### Standalone Tools (Always Call)
- **`session_init`** - Initialize session with workspace detection + context
- **`context_smart`** - Semantic search for relevant context (CALL EVERY MESSAGE, including immediately after `session_init`)
### Domain Tools (Use action/mode parameter)
| Domain | Actions/Modes | Example |
|--------|---------------|---------|
| **`search`** | mode: semantic, hybrid, keyword, pattern, exhaustive, refactor | `search(mode="hybrid", query="auth implementation", limit=3)` |
| **`session`** | action: capture, capture_lesson, get_lessons, recall, remember, user_context, summary, compress, delta, smart_search, decision_trace | `session(action="capture", event_type="decision", title="Use JWT", content="...")` |
| **`memory`** | action: create_event, get_event, update_event, delete_event, list_events, distill_event, create_node, get_node, update_node, delete_node, list_nodes, supersede_node, search, decisions, timeline, summary | `memory(action="list_events", limit=10)` |
| **`graph`** | action: dependencies, impact, call_path, related, path, decisions, ingest, circular_dependencies, unused_code, contradictions | `graph(action="impact", symbol_name="AuthService")` |
| **`project`** | action: list, get, create, update, index, overview, statistics, files, index_status, ingest_local | `project(action="statistics")` |
| **`workspace`** | action: list, get, associate, bootstrap | `workspace(action="list")` |
| **`reminder`** | action: list, active, create, snooze, complete, dismiss | `reminder(action="active")` |
| **`integration`** | provider: slack/github/all; action: status, search, stats, activity, contributors, knowledge, summary, channels, discussions, sync_users, repos, issues | `integration(provider="github", action="search", query="...")` |
| **`help`** | action: tools, auth, version, editor_rules, enable_bundle | `help(action="tools")` |
---
### Why context_smart is Required (Even After session_init)
**Common mistake:** "session_init already gave me context, I don't need context_smart"
**This is WRONG. Here's why:**
- `session_init` returns the last ~10 items **BY TIME** (chronological)
- `context_smart` **SEARCHES** for items **RELEVANT to THIS message** (semantic)
**Example failure:**
- User asks: "how should I implement authentication?"
- Auth decisions were made 20 conversations ago
- `session_init` won't have it (too old, not in recent 10)
- `context_smart` FINDS it via semantic search
**Without context_smart, you WILL miss relevant older context.**
---
### Search & Code Intelligence (ContextStream-first)
⚠️ **STOP: Before using Search/Glob/Grep/Read/Explore** → Call `search(mode="hybrid")` FIRST. Use local tools ONLY if ContextStream returns 0 results.
**❌ WRONG workflow (wastes tokens, slow):**
Grep "function" → Read file1.ts → Read file2.ts → Read file3.ts → ようやく理解
**✅ CORRECT workflow (fast, complete):**
search(mode="hybrid", query="function implementation") → 完了 (結果にコンテキストが含まれる)
**Why?** ContextStream search returns semantic matches + context + file locations in ONE call. Local tools require multiple round-trips.
**Search Mode Selection:**
| Need | Mode | Example |
|------|------|---------|
| Find code by meaning | `hybrid` | "authentication logic", "error handling" |
| Exact string/symbol | `keyword` | "UserAuthService", "API_KEY" |
| File patterns | `pattern` | "*.sql", "test_*.py" |
| ALL matches (grep-like) | `exhaustive` | "TODO", "FIXME" (find all occurrences) |
| Symbol renaming | `refactor` | "oldFunctionName" (word-boundary matching) |
| Conceptual search | `semantic` | "how does caching work" |
**Token Efficiency:** Use `output_format` to reduce response size:
- `full` (default): Full content for understanding code
- `paths`: File paths only (80% token savings) - use for file listings
- `minimal`: Compact format (60% savings) - use for refactoring
- `count`: Match counts only (90% savings) - use for quick checks
---
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
---
### Plans & Tasks
When user asks to create a plan or implementation roadmap:
1. Create plan: `session(action="capture_plan", title="Plan Title", description="...", goals=["goal1", "goal2"], steps=[{id: "1", title: "Step 1", order: 1}, ...])`
2. Get plan_id from response, then create tasks: `memory(action="create_task", title="Task Title", plan_id="<plan_id>", priority="high|medium|low", description="...")`
To manage existing plans/tasks:
- List plans: `session(action="list_plans")`
- Get plan with tasks: `session(action="get_plan", plan_id="<uuid>", include_tasks=true)`
- List tasks: `memory(action="list_tasks", plan_id="<uuid>")` or `memory(action="list_tasks")` for all
- Update task status: `memory(action="update_task", task_id="<uuid>", task_status="pending|in_progress|completed|blocked")`
---
### Rules Update Notices
- If you see **[RULES_NOTICE]**, update rules via `generate_rules(overwrite_existing=true)` (preserves custom rules).
- If you see **[VERSION_NOTICE]**, tell the user to update MCP using the provided command.
See full documentation: https://contextstream.io/docs/mcp/tools
Cline
プロジェクトルートに .clinerules ファイルを作成するか、.clinerules/ フォルダを使用します。
terminal · .clinerules (標準)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x (Hooks Enforced)
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
### Required Every Message
| Action | Tool Call |
|--------|-----------|
| **1st message** | `session_init(folder_path="<cwd>", context_hint="<msg>")` then `context_smart(...)` |
| **2nd+ messages** | `context_smart(user_message="<msg>", format="minified", max_tokens=400)` |
| **Code search** | `search(mode="hybrid", query="...")` — BEFORE any local tools |
| **Save decisions** | `session(action="capture", event_type="decision", ...)` |
### Search Modes
| Mode | Use Case |
|------|----------|
| `hybrid` | General code search (default) |
| `keyword` | Exact symbol/string match |
| `exhaustive` | Find ALL matches (grep-like) |
| `semantic` | Conceptual questions |
### Why ContextStream First?
❌ **WRONG:** `Grep → Read → Read → Read` (4+ tool calls, slow)
✅ **CORRECT:** `search(mode="hybrid")` (1 call, returns context)
ContextStream search is **indexed** and returns semantic matches + context in ONE call.
### Quick Reference
| Tool | Example |
|------|---------|
| `search` | `search(mode="hybrid", query="auth", limit=3)` |
| `session` | `session(action="capture", event_type="decision", title="...", content="...")` |
| `memory` | `memory(action="list_events", limit=10)` |
| `graph` | `graph(action="dependencies", file_path="...")` |
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
### Plans & Tasks
When user asks for a plan, use ContextStream (not EnterPlanMode):
1. `session(action="capture_plan", title="...", steps=[...])`
2. `memory(action="create_task", title="...", plan_id="<id>")`
Full docs: https://contextstream.io/docs/mcp/tools
拡張ルールを表示 (詳細)
terminal · .clinerules (拡張)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x Integration (Enhanced)
You have access to ContextStream MCP tools for persistent memory and context.
v0.4.x uses **~11 consolidated domain tools** for ~75% token reduction vs previous versions.
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
## TL;DR - REQUIRED EVERY MESSAGE
| Message | What to Call |
|---------|--------------|
| **1st message** | `session_init(folder_path="...", context_hint="<user's message>")`, then `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **2nd+ messages** | `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **🔍 ANY code search** | `search(mode="hybrid", query="...")` — ALWAYS before Glob/Grep/Search/Read |
| **Before risky/non-trivial work** | `session(action="get_lessons", query="<topic>")` |
| **After completing task** | `session(action="capture", event_type="decision", ...)` - MUST capture |
| **User frustration/correction** | `session(action="capture_lesson", ...)` - MUST capture lessons |
**NO EXCEPTIONS.** Do not skip even if you think you have enough context.
**First message rule:** After `session_init`, always call `context_smart` before any other tool or response.
**Context Pack (Pro+):** If enabled, use `context_smart(..., mode="pack", distill=true)` for code/file queries. If unavailable or disabled, omit `mode` and proceed with standard `context_smart` (the API will fall back).
**Tool naming:** Use the exact tool names exposed by your MCP client. Claude Code typically uses `mcp__<server>__<tool>` where `<server>` matches your MCP config (often `contextstream`). If a tool call fails with "No such tool available", refresh rules and match the tool list.
---
## Consolidated Domain Tools Architecture
v0.4.x consolidates ~58 individual tools into ~11 domain tools with action/mode dispatch:
### Standalone Tools (Always Call)
- **`session_init`** - Initialize session with workspace detection + context
- **`context_smart`** - Semantic search for relevant context (CALL EVERY MESSAGE, including immediately after `session_init`)
### Domain Tools (Use action/mode parameter)
| Domain | Actions/Modes | Example |
|--------|---------------|---------|
| **`search`** | mode: semantic, hybrid, keyword, pattern, exhaustive, refactor | `search(mode="hybrid", query="auth implementation", limit=3)` |
| **`session`** | action: capture, capture_lesson, get_lessons, recall, remember, user_context, summary, compress, delta, smart_search, decision_trace | `session(action="capture", event_type="decision", title="Use JWT", content="...")` |
| **`memory`** | action: create_event, get_event, update_event, delete_event, list_events, distill_event, create_node, get_node, update_node, delete_node, list_nodes, supersede_node, search, decisions, timeline, summary | `memory(action="list_events", limit=10)` |
| **`graph`** | action: dependencies, impact, call_path, related, path, decisions, ingest, circular_dependencies, unused_code, contradictions | `graph(action="impact", symbol_name="AuthService")` |
| **`project`** | action: list, get, create, update, index, overview, statistics, files, index_status, ingest_local | `project(action="statistics")` |
| **`workspace`** | action: list, get, associate, bootstrap | `workspace(action="list")` |
| **`reminder`** | action: list, active, create, snooze, complete, dismiss | `reminder(action="active")` |
| **`integration`** | provider: slack/github/all; action: status, search, stats, activity, contributors, knowledge, summary, channels, discussions, sync_users, repos, issues | `integration(provider="github", action="search", query="...")` |
| **`help`** | action: tools, auth, version, editor_rules, enable_bundle | `help(action="tools")` |
---
### Why context_smart is Required (Even After session_init)
**Common mistake:** "session_init already gave me context, I don't need context_smart"
**This is WRONG. Here's why:**
- `session_init` returns the last ~10 items **BY TIME** (chronological)
- `context_smart` **SEARCHES** for items **RELEVANT to THIS message** (semantic)
**Example failure:**
- User asks: "how should I implement authentication?"
- Auth decisions were made 20 conversations ago
- `session_init` won't have it (too old, not in recent 10)
- `context_smart` FINDS it via semantic search
**Without context_smart, you WILL miss relevant older context.**
---
### Search & Code Intelligence (ContextStream-first)
⚠️ **STOP: Before using Search/Glob/Grep/Read/Explore** → Call `search(mode="hybrid")` FIRST. Use local tools ONLY if ContextStream returns 0 results.
**❌ WRONG workflow (wastes tokens, slow):**
Grep "function" → Read file1.ts → Read file2.ts → Read file3.ts → ようやく理解
**✅ CORRECT workflow (fast, complete):**
search(mode="hybrid", query="function implementation") → 完了 (結果にコンテキストが含まれる)
**Why?** ContextStream search returns semantic matches + context + file locations in ONE call. Local tools require multiple round-trips.
**Search Mode Selection:**
| Need | Mode | Example |
|------|------|---------|
| Find code by meaning | `hybrid` | "authentication logic", "error handling" |
| Exact string/symbol | `keyword` | "UserAuthService", "API_KEY" |
| File patterns | `pattern` | "*.sql", "test_*.py" |
| ALL matches (grep-like) | `exhaustive` | "TODO", "FIXME" (find all occurrences) |
| Symbol renaming | `refactor` | "oldFunctionName" (word-boundary matching) |
| Conceptual search | `semantic` | "how does caching work" |
**Token Efficiency:** Use `output_format` to reduce response size:
- `full` (default): Full content for understanding code
- `paths`: File paths only (80% token savings) - use for file listings
- `minimal`: Compact format (60% savings) - use for refactoring
- `count`: Match counts only (90% savings) - use for quick checks
---
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
---
### Plans & Tasks
When user asks to create a plan or implementation roadmap:
1. Create plan: `session(action="capture_plan", title="Plan Title", description="...", goals=["goal1", "goal2"], steps=[{id: "1", title: "Step 1", order: 1}, ...])`
2. Get plan_id from response, then create tasks: `memory(action="create_task", title="Task Title", plan_id="<plan_id>", priority="high|medium|low", description="...")`
To manage existing plans/tasks:
- List plans: `session(action="list_plans")`
- Get plan with tasks: `session(action="get_plan", plan_id="<uuid>", include_tasks=true)`
- List tasks: `memory(action="list_tasks", plan_id="<uuid>")` or `memory(action="list_tasks")` for all
- Update task status: `memory(action="update_task", task_id="<uuid>", task_status="pending|in_progress|completed|blocked")`
---
### Rules Update Notices
- If you see **[RULES_NOTICE]**, update rules via `generate_rules(overwrite_existing=true)` (preserves custom rules).
- If you see **[VERSION_NOTICE]**, tell the user to update MCP using the provided command.
See full documentation: https://contextstream.io/docs/mcp/tools
Roo Code
.roo/rules/contextstream.md ファイルを作成するか、.roo/rules/ フォルダを使用します。
terminal · .roo/rules/contextstream.md (標準)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x (Hooks Enforced)
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
### Required Every Message
| Action | Tool Call |
|--------|-----------|
| **1st message** | `session_init(folder_path="<cwd>", context_hint="<msg>")` then `context_smart(...)` |
| **2nd+ messages** | `context_smart(user_message="<msg>", format="minified", max_tokens=400)` |
| **Code search** | `search(mode="hybrid", query="...")` — BEFORE any local tools |
| **Save decisions** | `session(action="capture", event_type="decision", ...)` |
### Search Modes
| Mode | Use Case |
|------|----------|
| `hybrid` | General code search (default) |
| `keyword` | Exact symbol/string match |
| `exhaustive` | Find ALL matches (grep-like) |
| `semantic` | Conceptual questions |
### Why ContextStream First?
❌ **WRONG:** `Grep → Read → Read → Read` (4+ tool calls, slow)
✅ **CORRECT:** `search(mode="hybrid")` (1 call, returns context)
ContextStream search is **indexed** and returns semantic matches + context in ONE call.
### Quick Reference
| Tool | Example |
|------|---------|
| `search` | `search(mode="hybrid", query="auth", limit=3)` |
| `session` | `session(action="capture", event_type="decision", title="...", content="...")` |
| `memory` | `memory(action="list_events", limit=10)` |
| `graph` | `graph(action="dependencies", file_path="...")` |
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
### Plans & Tasks
When user asks for a plan, use ContextStream (not EnterPlanMode):
1. `session(action="capture_plan", title="...", steps=[...])`
2. `memory(action="create_task", title="...", plan_id="<id>")`
Full docs: https://contextstream.io/docs/mcp/tools
拡張ルールを表示 (詳細)
terminal · .roo/rules/contextstream.md (拡張)
## 🚨 MANDATORY RULE: DISPLAY AT START OF EVERY RESPONSE 🚨
<contextstream_rules>
1. **SEARCH FIRST:** Before Glob/Grep/Search/Read → `search(mode="hybrid", query="...")` FIRST
2. **EVERY MESSAGE:** Call `context_smart(user_message="...")` at start of response
3. **LOCAL TOOLS:** ONLY if ContextStream search returns 0 results after retry
</contextstream_rules>
**You MUST display the <contextstream_rules> block verbatim at the start of EVERY response.**
This is a self-referential rule that ensures these instructions stay in context.
---
## ContextStream v0.4.x Integration (Enhanced)
You have access to ContextStream MCP tools for persistent memory and context.
v0.4.x uses **~11 consolidated domain tools** for ~75% token reduction vs previous versions.
Rules Version: 0.4.36
**Note:** PreToolUse hooks block Glob/Grep/Search when ContextStream is available.
## TL;DR - REQUIRED EVERY MESSAGE
| Message | What to Call |
|---------|--------------|
| **1st message** | `session_init(folder_path="...", context_hint="<user's message>")`, then `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **2nd+ messages** | `context_smart(user_message="<user's message>", format="minified", max_tokens=400)` |
| **🔍 ANY code search** | `search(mode="hybrid", query="...")` — ALWAYS before Glob/Grep/Search/Read |
| **Before risky/non-trivial work** | `session(action="get_lessons", query="<topic>")` |
| **After completing task** | `session(action="capture", event_type="decision", ...)` - MUST capture |
| **User frustration/correction** | `session(action="capture_lesson", ...)` - MUST capture lessons |
**NO EXCEPTIONS.** Do not skip even if you think you have enough context.
**First message rule:** After `session_init`, always call `context_smart` before any other tool or response.
**Context Pack (Pro+):** If enabled, use `context_smart(..., mode="pack", distill=true)` for code/file queries. If unavailable or disabled, omit `mode` and proceed with standard `context_smart` (the API will fall back).
**Tool naming:** Use the exact tool names exposed by your MCP client. Claude Code typically uses `mcp__<server>__<tool>` where `<server>` matches your MCP config (often `contextstream`). If a tool call fails with "No such tool available", refresh rules and match the tool list.
---
## Consolidated Domain Tools Architecture
v0.4.x consolidates ~58 individual tools into ~11 domain tools with action/mode dispatch:
### Standalone Tools (Always Call)
- **`session_init`** - Initialize session with workspace detection + context
- **`context_smart`** - Semantic search for relevant context (CALL EVERY MESSAGE, including immediately after `session_init`)
### Domain Tools (Use action/mode parameter)
| Domain | Actions/Modes | Example |
|--------|---------------|---------|
| **`search`** | mode: semantic, hybrid, keyword, pattern, exhaustive, refactor | `search(mode="hybrid", query="auth implementation", limit=3)` |
| **`session`** | action: capture, capture_lesson, get_lessons, recall, remember, user_context, summary, compress, delta, smart_search, decision_trace | `session(action="capture", event_type="decision", title="Use JWT", content="...")` |
| **`memory`** | action: create_event, get_event, update_event, delete_event, list_events, distill_event, create_node, get_node, update_node, delete_node, list_nodes, supersede_node, search, decisions, timeline, summary | `memory(action="list_events", limit=10)` |
| **`graph`** | action: dependencies, impact, call_path, related, path, decisions, ingest, circular_dependencies, unused_code, contradictions | `graph(action="impact", symbol_name="AuthService")` |
| **`project`** | action: list, get, create, update, index, overview, statistics, files, index_status, ingest_local | `project(action="statistics")` |
| **`workspace`** | action: list, get, associate, bootstrap | `workspace(action="list")` |
| **`reminder`** | action: list, active, create, snooze, complete, dismiss | `reminder(action="active")` |
| **`integration`** | provider: slack/github/all; action: status, search, stats, activity, contributors, knowledge, summary, channels, discussions, sync_users, repos, issues | `integration(provider="github", action="search", query="...")` |
| **`help`** | action: tools, auth, version, editor_rules, enable_bundle | `help(action="tools")` |
---
### Why context_smart is Required (Even After session_init)
**Common mistake:** "session_init already gave me context, I don't need context_smart"
**This is WRONG. Here's why:**
- `session_init` returns the last ~10 items **BY TIME** (chronological)
- `context_smart` **SEARCHES** for items **RELEVANT to THIS message** (semantic)
**Example failure:**
- User asks: "how should I implement authentication?"
- Auth decisions were made 20 conversations ago
- `session_init` won't have it (too old, not in recent 10)
- `context_smart` FINDS it via semantic search
**Without context_smart, you WILL miss relevant older context.**
---
### Search & Code Intelligence (ContextStream-first)
⚠️ **STOP: Before using Search/Glob/Grep/Read/Explore** → Call `search(mode="hybrid")` FIRST. Use local tools ONLY if ContextStream returns 0 results.
**❌ WRONG workflow (wastes tokens, slow):**
Grep "function" → Read file1.ts → Read file2.ts → Read file3.ts → ようやく理解
**✅ CORRECT workflow (fast, complete):**
search(mode="hybrid", query="function implementation") → 完了 (結果にコンテキストが含まれる)
**Why?** ContextStream search returns semantic matches + context + file locations in ONE call. Local tools require multiple round-trips.
**Search Mode Selection:**
| Need | Mode | Example |
|------|------|---------|
| Find code by meaning | `hybrid` | "authentication logic", "error handling" |
| Exact string/symbol | `keyword` | "UserAuthService", "API_KEY" |
| File patterns | `pattern` | "*.sql", "test_*.py" |
| ALL matches (grep-like) | `exhaustive` | "TODO", "FIXME" (find all occurrences) |
| Symbol renaming | `refactor` | "oldFunctionName" (word-boundary matching) |
| Conceptual search | `semantic` | "how does caching work" |
**Token Efficiency:** Use `output_format` to reduce response size:
- `full` (default): Full content for understanding code
- `paths`: File paths only (80% token savings) - use for file listings
- `minimal`: Compact format (60% savings) - use for refactoring
- `count`: Match counts only (90% savings) - use for quick checks
---
### Lessons (Past Mistakes)
- After `session_init`: Check for `lessons` field and apply before work
- Before risky work: `session(action="get_lessons", query="<topic>")`
- On mistakes: `session(action="capture_lesson", title="...", trigger="...", impact="...", prevention="...")`
---
### Plans & Tasks
When user asks to create a plan or implementation roadmap:
1. Create plan: `session(action="capture_plan", title="Plan Title", description="...", goals=["goal1", "goal2"], steps=[{id: "1", title: "Step 1", order: 1}, ...])`
2. Get plan_id from response, then create tasks: `memory(action="create_task", title="Task Title", plan_id="<plan_id>", priority="high|medium|low", description="...")`
To manage existing plans/tasks:
- List plans: `session(action="list_plans")`
- Get plan with tasks: `session(action="get_plan", plan_id="<uuid>", include_tasks=true)`
- List tasks: `memory(action="list_tasks", plan_id="<uuid>")` or `memory(action="list_tasks")` for all
- Update task status: `memory(action="update_task", task_id="<uuid>", task_status="pending|in_progress|completed|blocked")`
---
### Rules Update Notices
- If you see **[RULES_NOTICE]**, update rules via `generate_rules(overwrite_existing=true)` (preserves custom rules).
- If you see **[VERSION_NOTICE]**, tell the user to update MCP using the provided command.
See full documentation: https://contextstream.io/docs/mcp/tools
ルールの自動生成
AIに次のように依頼して、これらのルールを自動生成することもできます: "generate_rulesを使用して、このプロジェクトのContextStreamルールを作成してください"
レッスンシステム
教訓学習システム
失敗から学ぶ — 同じ過ちを繰り返さない
教訓システムは、ミス、修正、ユーザーの不満をキャプチャし、AIアシスタントが同じエラーを繰り返さないようにします。教訓は、関連性がある場合に session_init および context_smart の応答で自動的に表示されます。
教訓をキャプチャするタイミング
以下の状況が発生した場合、教訓は自動的にキャプチャされるべきです。
| トリガー | 例 | 重大度 |
|---|---|---|
| 本番環境の問題 | "その変更のせいでサイトがダウンしています" | critical |
| ユーザーの不満 | "違う！それをしないでと言ったでしょ"、"WTF"、大文字 | high |
| 修正 | "それは間違っています。こうすべきです..."、"これを修正して" | medium |
| 破壊的変更 | "これでテストが壊れました"、"ビルドが失敗しました" | medium/high |
| 好みの表明 | "私はこの方法が好きです"、"常にXの代わりにYをしてください" | low |
教訓フィールドの説明
| フィールド | 説明 | 例 |
|---|---|---|
| title | 覚えておくべきこと (命令形) | "プッシュする前に必ずアセットをgitで確認する" |
| severity | critical, high, medium, low | 本番環境の問題には"critical" |
| category | workflow, code_quality, verification, communication, project_specific | "workflow" |
| trigger | 問題を引き起こしたアクション | "コミットせずに画像を参照するコードをプッシュした" |
| impact | 何が問題だったか | "本番環境で404エラー - ランディングページが壊れた" |
| prevention | 今後どのように防ぐか | "プッシュする前にgit statusで追跡されていないファイルを確認する" |
| keywords | 将来のコンテキストでマッチングするためのキーワード | ["git", "images", "assets", "push"] |
完全な例
terminal · session_capture_lesson
// User says: "OH COME ON! You pushed the code but the images are missing
// and now the production site shows broken images!"
session_capture_lesson({
title: "Always verify assets in git before pushing code references",
severity: "critical",
category: "workflow",
trigger: "Pushed code referencing /screenshots/*.png without committing images",
impact: "Production 404 errors - broken landing page with missing images",
prevention: "Run 'git status' to check untracked files before pushing code that references static assets",
keywords: ["git", "images", "assets", "push", "404", "static", "screenshots"]
})
教訓の表示方法
キャプチャされた教訓は、関連性がある場合に将来のセッションで自動的に返されます。
session_init
このワークスペースからの高および重大な重大度の教訓がセッション初期化に含まれ、AIが同じミスを犯す前に警告します。
context_smart
AIがコンテキストを要求すると、クエリキーワードに一致する教訓が含まれます。例: "git push"について尋ねると、"git"または"push"キーワードを持つ教訓が表示されます。
教訓の取得
session_get_lessons を使用して教訓を取得およびフィルタリングします。
terminal · session_get_lessons 例
// Get all critical lessons
session_get_lessons({ severity: "critical" })
// Get workflow lessons
session_get_lessons({ category: "workflow" })
// Search for relevant lessons
session_get_lessons({ query: "git push images" })
// Combine filters
session_get_lessons({
category: "verification",
severity: "high",
limit: 5
})
プロのヒント: エディタにルールを追加して（上記の「エディタAIルール」セクションを参照）、ユーザーが不満や修正を表明したときに自動的に教訓を取得するようにしましょう。これにより、繰り返し発生するミスを防ぐナレッジベースが構築されます。
ツールカタログ
MCPツール
完全なMCPツールリファレンス（PROバッジと一般的な使用例）をご覧ください。
MCPツールリファレンスを表示統合ツールセットは、従来の細分化されたツールと比較してトークンを約75%削減する約11のドメインツールを使用します。詳細を見る
使用例
質問例
接続すると、AIアシスタントに次のような質問ができます：
「データベースにPostgreSQLを使用することに決めたことを覚えておいて」
「認証に関する以前の決定事項は何でしたか？」
「コードベースでAPIレート制限の処理方法を検索して」
「支払いシステムに関する関連コンテキストを表示して」
教訓の例
ユーザーが不満や修正を表明すると、AIは自動的に教訓を取得する必要があります：
ユーザーの発言
「ダメだ！テストを実行せずにプッシュしたせいで本番環境が壊れてる！」
→ AIが教訓を取得: 重要度: critical, カテゴリ: verification
ユーザーの発言
「それは間違いです。データベースのカラムには常にcamelCaseではなくsnake_caseを使用してください。」
→ AIが教訓を取得: 重要度: medium, カテゴリ: code_quality
ユーザーの発言
「TypeScriptのstrictモードが好みです。常に有効にしてください。」
→ AIが教訓を取得: 重要度: low, カテゴリ: project_specific
これらの教訓は、関連するコンテキストが要求された際に、将来のセッションで自動的に表示されます。
メンテナンス
最新の状態を保つ
最新の機能、バグ修正、改善を入手するには、MCPサーバーを定期的に更新してください：
terminal · terminal · macOS / Linux
curl -fsSL https://contextstream.io/scripts/mcp.sh | bash
terminal · powershell · Windows
irm https://contextstream.io/scripts/mcp.ps1 | iex
MCPサーバーは、新しいバージョンが利用可能になると自動的に警告します。更新後、AIツールを再起動して新しいバージョンを使用してください。
トラブルシューティング
うまく動作しない場合
MCPサーバーが起動しない
contextstream-mcp がインストールされ、PATHで利用可能であることを確認してください。 contextstream-mcp --version を手動で実行してエラーを確認してみてください。
認証エラー
APIキーが正しく、有効期限が切れていないことを確認してください。ContextStreamダッシュボードから新しいキーを生成できます。
ツールが表示されない
設定を変更した後、AIアプリケーションを再起動してください。アプリケーションログでMCP接続エラーを確認してください。
ワークスペースが見つからない（初回セットアップ時）
アカウントにまだワークスペースがない場合、ContextStreamはAIアシスタントにワークスペース名を尋ねるよう促します。現在のフォルダがプロジェクトとして作成されます。 workspace_bootstrap については、MCPツールを参照してください。
次のステップ
さらに探索する
チーム設定メンバーを招待し、コンテキストを共有します。詳細を見る学んだ教訓ミスを取得し、二度と繰り返さないようにします。詳細を見るメモリイベントメモリタイプについて学びます。詳細を見るセマンティック検索意味で検索します。詳細を見る
{"@context":"https://schema.org","@type":"Organization","name":"ContextStream","url":"https://contextstream.io","logo":"https://contextstream.io/logo.png","description":"Give your AI applications infinite context. Unify code, documentation, and workspace history into a single, queryable intelligence layer with semantic search and knowledge graphs.","sameAs":["https://twitter.com/contextstream","https://github.com/contextstream"]}
{"@context":"https://schema.org","@type":"SoftwareApplication","name":"ContextStream","applicationCategory":"DeveloperApplication","operatingSystem":"Any","description":"Give your AI applications infinite context. Unify code, documentation, and workspace history into a single, queryable intelligence layer with semantic search and knowledge graphs.","offers":{"@type":"Offer","price":"20","priceCurrency":"USD","description":"Pro plan includes a 5-day free trial"},"aggregateRating":{"@type":"AggregateRating","ratingValue":"5","ratingCount":"10"}}