setup

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Setup Repo-Native Alignment MCP

安装配置Repo-Native Alignment (RNA) MCP

Install the RNA MCP server for aim-conditioned code intelligence.

Execute these steps in order. Do not stop between steps or ask for confirmation -- run the full sequence automatically.

安装用于目标导向型代码智能的RNA MCP服务器。

请按顺序执行以下步骤。步骤间不要停止或请求确认——自动运行完整流程。

Step 1: Check if already installed

步骤1：检查是否已安装

bash

which repo-native-alignment 2>/dev/null

If found, skip to Step 3. If not found, proceed to Step 2.

bash

which repo-native-alignment 2>/dev/null

如果已找到，跳至步骤3。如果未找到，继续步骤2。

Step 2: Download the binary

步骤2：下载二进制文件

Detect the platform and chip, then download to

~/.cargo/bin/

(already on PATH for Rust users):

bash

OS=$(uname -s)
ARCH=$(uname -m)
CHIP=$(sysctl -n machdep.cpu.brand_string 2>/dev/null || echo "")
mkdir -p ~/.cargo/bin

If macOS ARM M2+ (

Darwin

arm64

+ brand_string contains "M2", "M3", or "M4"):

bash

curl -L https://github.com/open-horizon-labs/repo-native-alignment/releases/latest/download/repo-native-alignment-darwin-arm64-fast.tar.gz | tar xz -C ~/.cargo/bin

If macOS ARM (M1) (

Darwin

arm64

bash

curl -L https://github.com/open-horizon-labs/repo-native-alignment/releases/latest/download/repo-native-alignment-darwin-arm64.tar.gz | tar xz -C ~/.cargo/bin

If Linux x86_64 (

Linux

x86_64

bash

curl -L https://github.com/open-horizon-labs/repo-native-alignment/releases/latest/download/repo-native-alignment-linux-x86_64.tar.gz | tar xz -C ~/.cargo/bin

If none of the above match: Tell the user their platform is not yet supported. They can build from source:

bash

cargo install --locked --git https://github.com/open-horizon-labs/repo-native-alignment

~/.cargo/bin

is not on PATH (no Rust toolchain installed), tell the user to add it:

export PATH="$HOME/.cargo/bin:$PATH"

检测平台和芯片，然后下载至

~/.cargo/bin/

（Rust用户的该路径已在PATH中）：

bash

OS=$(uname -s)
ARCH=$(uname -m)
CHIP=$(sysctl -n machdep.cpu.brand_string 2>/dev/null || echo "")
mkdir -p ~/.cargo/bin

如果是macOS ARM M2+（

Darwin

arm64

+ 品牌字符串包含"M2"、"M3"或"M4"）：

bash

curl -L https://github.com/open-horizon-labs/repo-native-alignment/releases/latest/download/repo-native-alignment-darwin-arm64-fast.tar.gz | tar xz -C ~/.cargo/bin

如果是macOS ARM (M1)（

Darwin

arm64

）：

bash

curl -L https://github.com/open-horizon-labs/repo-native-alignment/releases/latest/download/repo-native-alignment-darwin-arm64.tar.gz | tar xz -C ~/.cargo/bin

如果是Linux x86_64（

Linux

x86_64

）：

bash

curl -L https://github.com/open-horizon-labs/repo-native-alignment/releases/latest/download/repo-native-alignment-linux-x86_64.tar.gz | tar xz -C ~/.cargo/bin

如果以上都不匹配：告知用户其平台暂不支持。他们可以从源码构建：

bash

cargo install --locked --git https://github.com/open-horizon-labs/repo-native-alignment

如果

~/.cargo/bin

不在PATH中（未安装Rust工具链），告知用户添加该路径：

export PATH="$HOME/.cargo/bin:$PATH"

Step 3: Configure the MCP server

步骤3：配置MCP服务器

RNA is a per-project MCP server (it indexes the repo it's pointed at).

Check if

.mcp.json

exists in the project root and already contains an

rna-mcp

entry. If it does, skip this step.

If the agent supports

claude mcp add

(Claude Code):

bash

claude mcp add rna-mcp --scope project -- repo-native-alignment --repo .

Otherwise, create or update

.mcp.json

in the project root with:

json

{
  "mcpServers": {
    "rna-mcp": {
      "command": "repo-native-alignment",
      "args": ["--repo", "."]
    }
  }
}

.mcp.json

already exists with other servers, merge the

rna-mcp

entry into the existing

mcpServers

object -- do not overwrite the file.

RNA是一个基于项目的MCP服务器（它会为指向的代码仓库建立索引）。

检查项目根目录下是否存在

.mcp.json

且已包含

rna-mcp

条目。如果存在，跳过此步骤。

如果Agent支持

claude mcp add

（Claude Code）：

bash

claude mcp add rna-mcp --scope project -- repo-native-alignment --repo .

否则，在项目根目录下创建或更新

.mcp.json

，内容如下：

json

{
  "mcpServers": {
    "rna-mcp": {
      "command": "repo-native-alignment",
      "args": ["--repo", "."]
    }
  }
}

如果

.mcp.json

已存在且包含其他服务器条目，请将

rna-mcp

条目合并到现有的

mcpServers

对象中——不要覆盖整个文件。

Step 4: Pre-warm the code index

步骤4：预加载代码索引

Run a full scan to build the code index before the MCP server starts. This avoids cold-start latency on the first tool call:

bash

repo-native-alignment scan --repo . --full

This builds the full pipeline (scan, extract, embed, LSP enrich, graph) and caches results in

.oh/.cache/lance/

. The MCP server reuses this cache on startup -- if no files changed, graph loads in seconds with zero re-extraction. Subsequent scans are incremental.

Without this step, the MCP server pre-warms the graph automatically at startup, but the first tool call may need to wait for that to complete. Pre-building ensures instant readiness.

在MCP服务器启动前运行完整扫描以构建代码索引。这样可以避免首次工具调用时的冷启动延迟：

bash

repo-native-alignment scan --repo . --full

此命令会构建完整流程（扫描、提取、嵌入、LSP增强、图谱）并将结果缓存到

.oh/.cache/lance/

中。MCP服务器启动时会复用该缓存——如果文件未发生变化，图谱可在数秒内加载完成，无需重新提取。后续扫描为增量式。

如果跳过此步骤，MCP服务器会在启动时自动预加载图谱，但首次工具调用可能需要等待该过程完成。提前构建可确保立即就绪。

Step 5: Update AGENTS.md with tool guidance

步骤5：更新AGENTS.md添加工具使用指南

If AGENTS.md exists in the project root, check if it already contains

<!-- RNA MCP tool guidance -->

. If it already has this marker, skip this step.

If AGENTS.md exists but lacks the marker, append this block:

markdown

<!-- RNA MCP tool guidance -->

如果项目根目录下存在AGENTS.md，检查其是否已包含

<!-- RNA MCP tool guidance -->

标记。如果已包含，跳过此步骤。

如果AGENTS.md存在但缺少该标记，追加以下内容块：

markdown

<!-- RNA MCP tool guidance -->

Code Exploration (use RNA MCP tools, not grep/Read)

代码探索（使用RNA MCP工具，而非grep/Read）

Instead of...	Use this MCP tool
`Grep` for symbol names	`search_symbols(query, kind, language, file)`
`Read` to trace function calls	`graph_query(node_id, mode: "neighbors")`
`Grep` for "who calls X"	`graph_query(node_id, mode: "impact")`
`Read` to find .oh/ artifacts	`oh_search_context(query)`
`Bash` with `grep -rn`	`search_symbols` or `oh_search_context`
Recording learnings/signals	Write to `.oh/metis/` , `.oh/signals/` , `.oh/guardrails/` (YAML frontmatter + markdown)
Searching git history	`oh_search_context(query)` -- returns hash; use `git show <hash>` via Bash for diffs


If AGENTS.md does not exist, create it with the tool guidance block as the initial content.

替代...	使用此MCP工具
`Grep` 查找符号名称	`search_symbols(query, kind, language, file)`
`Read` 追踪函数调用	`graph_query(node_id, mode: "neighbors")`
`Grep` 查找"谁调用了X"	`graph_query(node_id, mode: "impact")`
`Read` 查找.oh/ 产物	`oh_search_context(query)`
`Bash` 搭配 `grep -rn`	`search_symbols` 或 `oh_search_context`
记录学习成果/信号	写入 `.oh/metis/` , `.oh/signals/` , `.oh/guardrails/` （YAML前置元数据 + markdown）
搜索Git历史	`oh_search_context(query)` -- 返回哈希值；通过Bash使用 `git show <hash>` 查看差异


如果AGENTS.md不存在，创建该文件并将上述工具指南块作为初始内容。

Step 6: Inform the user

步骤6：告知用户

Tell the user:

Setup is complete
They may need to restart their agent/IDE for the MCP server to load
After restart, RNA MCP tools will be available:
- ```
oh_get_context
```
  - Business context in one call
- ```
oh_search_context
```
  - Semantic search over outcomes, code, commits
- ```
search_symbols
```
  - Multi-language symbol search with graph edges
- ```
graph_query
```
  - Impact analysis, neighbor traversal, reachability
- ```
outcome_progress
```
  - Structural outcome-to-code joins
- ```
list_roots
```
  - Workspace root management

告知用户：

安装配置已完成
他们可能需要重启Agent/IDE以加载MCP服务器
重启后，RNA MCP工具将可用：
- ```
oh_get_context
```
  - 一键获取业务上下文
- ```
oh_search_context
```
  - 对产出物、代码、提交记录进行语义搜索
- ```
search_symbols
```
  - 带图谱关联的多语言符号搜索
- ```
graph_query
```
  - 影响分析、关联节点遍历、可达性分析
- ```
outcome_progress
```
  - 结构化产出物与代码关联
- ```
list_roots
```
  - 工作区根目录管理