### Run the example via CLI

Source: https://docs.ratel.sh/docs/core/framework-integrations

Commands to install dependencies and execute the example project from the repository.

```bash
export OPENAI_API_KEY=sk-...
pnpm install
pnpm -F @ratel-ai/example-ai-sdk start "send an email to alice@example.com saying ship it"
```

--------------------------------

### Install Skill Hook with User Scope

Source: https://docs.ratel.sh/docs/local/managing-skills

Example command to install a skill hook targeting the user settings file.

```bash
ratel-mcp skill install-hook --scope user
```

--------------------------------

### Install all skills

Source: https://docs.ratel.sh/docs/core/skills-suite

Installs all nine skills into the current project directory.

```bash
npx skills add ratel-ai/skills -y --all
```

--------------------------------

### Install and Run Ratel Local

Source: https://docs.ratel.sh/docs/local

Install the package globally or execute it directly using npx.

```bash
pnpm add -g @ratel-ai/mcp-server

# or skip the install and run it on the fly
npx -y @ratel-ai/mcp-server --help
```

--------------------------------

### Install Ratel SDKs

Source: https://docs.ratel.sh/docs/core/sdks

Installation commands for Python and TypeScript environments.

```bash
pip install ratel-ai
# upstream MCP ingestion (register_mcp_server) needs the extra:
pip install 'ratel-ai[mcp]'
```

```bash
pnpm add @ratel-ai/sdk
# or
npm install @ratel-ai/sdk
```

--------------------------------

### Install from local directory

Source: https://docs.ratel.sh/docs/core/skills-suite

Used for local development when iterating on the skills suite itself.

```bash
git clone https://github.com/ratel-ai/skills
npx skills add ./skills
```

--------------------------------

### Quickstart integration

Source: https://docs.ratel.sh/docs/core/skills-suite

Command to run within a coding agent to initiate the Ratel integration plan.

```bash
Run npx skills add ratel-ai/skills --all and use the skills to integrate Ratel in this project.
```

--------------------------------

### Install Ratel Python SDK

Source: https://docs.ratel.sh/docs/core/sdks/python

Commands to install the base SDK and the optional MCP ingestion support.

```bash
pip install ratel-ai
# upstream MCP ingestion (register_mcp_server) needs the extra:
pip install 'ratel-ai[mcp]'
```

--------------------------------

### Quickstart commands for coding agents

Source: https://docs.ratel.sh/llms-full.txt

Commands to be pasted directly into a coding agent for assessment or integration.

```text
Run npx skills add ratel-ai/skills --all and use the skills to assess the agents in this codebase
```

```text
Run npx skills add ratel-ai/skills --all and use the skills to integrate Ratel in this project.
```

--------------------------------

### Install Ratel SDK dependencies

Source: https://docs.ratel.sh/docs/core/quickstart

Commands to initialize a project and install necessary packages for TypeScript and Python environments.

```bash
mkdir ratel-quickstart && cd ratel-quickstart
npm init -y
npm pkg set type=module
npm install @ratel-ai/sdk@0.4.0 ai@^6 @ai-sdk/openai@^3
npm install --save-dev tsx
```

```bash
mkdir ratel-quickstart && cd ratel-quickstart
python -m venv .venv
source .venv/bin/activate
python -m pip install "ratel-ai==0.4.0" "pydantic-ai>=1.0"
```

--------------------------------

### Quickstart assessment

Source: https://docs.ratel.sh/docs/core/skills-suite

Command to run within a coding agent to perform a static codebase assessment.

```bash
Run npx skills add ratel-ai/skills --all and use the skills to assess the agents in this codebase
```

--------------------------------

### Install a specific skill

Source: https://docs.ratel.sh/docs/core/skills-suite

Installs only the specified skill into the current project.

```bash
npx skills add ratel-ai/skills --skill ratel-assessment
```

--------------------------------

### Install Ratel SDK

Source: https://docs.ratel.sh/

Use npm to install the Ratel SDK for TypeScript projects.

```bash
$npm install @ratel-ai/sdk
```

--------------------------------

### Install OTLP Exporter for Python

Source: https://docs.ratel.sh/docs/core/telemetry

Install the required extra package to enable OTLP telemetry support in Python.

```bash
pip install 'ratel-ai[otlp]'
```

--------------------------------

### Invoke Tool Execution Example

Source: https://docs.ratel.sh/docs/core/capability-tools

Demonstrates the expected input format and the raw result returned upon successful tool execution.

```json
// invoke_tool({ "toolId": "fs__read_file", "args": { "path": "/tmp/x" } })
{ "contents": "contents of /tmp/x" }
```

--------------------------------

### Search Capabilities via MCP Client

Source: https://docs.ratel.sh/docs/local

Example prompt for an MCP client to query available tools.

```text
Call search_capabilities with the query "read a file" and show me the raw result.
```

--------------------------------

### Launch local browser UI

Source: https://docs.ratel.sh/docs/local

Commands to start the ephemeral browser-based UI for managing servers, OAuth, and skills.

```bash
ratel-mcp ui              # ephemeral 127.0.0.1 port, opens your browser
ratel-mcp ui --port 5731  # bind a specific port
ratel-mcp ui --no-open    # print the URL without launching a browser
```

--------------------------------

### Skill suggestion output

Source: https://docs.ratel.sh/docs/local/managing-skills

Example output format for a skill suggestion.

```text
auth-flows  (score 2.31)  How to add authentication: sessions, OAuth, form validation.
```

--------------------------------

### Skill list output

Source: https://docs.ratel.sh/docs/local/managing-skills

Example output showing the list of skills currently managed by Ratel.

```text
Ratel manages 2 skill(s):
  deploy-runbook  (from /Users/you/.claude/skills/deploy-runbook)
  debugging-checklist  (from /Users/you/.claude/skills/debugging-checklist)
```

--------------------------------

### Skill activation output

Source: https://docs.ratel.sh/docs/local/managing-skills

Example output when running the activation command in dry-run mode.

```text
would activate deploy-runbook
would activate debugging-checklist
```

--------------------------------

### init(options: InitOptions)

Source: https://docs.ratel.sh/docs/core/api/sdk-typescript

Initializes the telemetry client with OTLP configuration. You can provide an apiKey for automatic Bearer authentication or a custom endpoint and headers for advanced collector setups.

```APIDOC
## init(options: InitOptions)

### Description
Initializes the telemetry client. An explicit `endpoint` overrides the default `RATEL_URL`, and `apiKey` is automatically added as an `Authorization: Bearer` header.

### Parameters
- **apiKey** (string) - Optional - Ratel API key.
- **endpoint** (string) - Optional - Full OTLP traces URL (e.g., /v1/traces).
- **headers** (Record<string, string>) - Optional - Extra headers merged into the request.
- **serviceName** (string) - Optional - The service.name resource attribute.
```

--------------------------------

### Example MCP configuration file

Source: https://docs.ratel.sh/docs/local

The structure of the JSON configuration file for defining stdio and HTTP MCP servers.

```json
{
  "mcpServers": {
    "ev": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-everything"],
      "description": "filesystem & shell utilities"
    },
    "remote": {
      "type": "http",
      "url": "https://example.com/mcp",
      "headers": { "Authorization": "Bearer xyz" }
    }
  }
}
```

--------------------------------

### ratel-mcp skill install-hook

Source: https://docs.ratel.sh/docs/local/managing-skills

Installs a UserPromptSubmit hook to the specified settings file, enabling automatic skill preloading.

```APIDOC
## ratel-mcp skill install-hook

### Description
Adds a push-path hook to the configuration file that runs `skill preload-hook` on every prompt, allowing the agent to inject relevant skills.

### Parameters
- **--scope** (string) - Optional - Target configuration file: 'user' (~/.claude/settings.json) or 'project' (<projectRoot>/.claude/settings.json). Defaults to 'user'.
- **--yes** (boolean) - Optional - Skip the confirmation prompt during installation.

### Request Example
ratel-mcp skill install-hook --scope user
```

--------------------------------

### Run Skill Suggestion

Source: https://docs.ratel.sh/llms-full.txt

Example usage of the command to find skills relevant to a specific prompt within the current directory.

```bash
ratel-mcp skill suggest --prompt "build a login form" --cwd . --limit 3
```

--------------------------------

### Install Ratel SDK

Source: https://docs.ratel.sh/docs/core/sdks/typescript

Commands to add the Ratel SDK dependency to your project using pnpm or npm.

```bash
pnpm add @ratel-ai/sdk
# or
npm install @ratel-ai/sdk
```

--------------------------------

### Run Ratel MCP Server

Source: https://docs.ratel.sh/docs/local

Commands to start the server using explicit configuration files or automatic discovery modes.

```bash
ratel-mcp serve --config ~/.ratel/config.json
ratel-mcp serve ~/.ratel/config.json                 # positional form of the same thing
ratel-mcp serve --auto-config                        # resolve the user, project, and local configs automatically
ratel-mcp serve --auto-config --project-root /path/to/project  # pin project and local scopes
```

--------------------------------

### Verify Search Capabilities

Source: https://docs.ratel.sh/docs/local/quickstart

Example tool call to verify that Ratel search capabilities are functioning correctly.

```json
Call Ratel's search_capabilities tool with:
{"query":"look up current React framework documentation","topKTools":3,"topKSkills":1}
Return the raw result.
```

--------------------------------

### Install Ratel Local CLI

Source: https://docs.ratel.sh/docs/local/quickstart

Requires Node.js 20 or newer. Use the ratel-mcp binary for the import flow.

```bash
npm install --global @ratel-ai/mcp-server@0.4.0
ratel-mcp --version
```

--------------------------------

### Initialize Capability Tools

Source: https://docs.ratel.sh/docs/core/sdks/python

Create the search and invoke tools from a provided catalog.

```python
search = search_capabilities_tool(catalog)  # id == "search_capabilities"
invoke = invoke_tool_tool(catalog)          # id == "invoke_tool"
```

--------------------------------

### Install OTLP Exporter for TypeScript

Source: https://docs.ratel.sh/docs/core/telemetry

Install the telemetry package required for OTLP exporting in TypeScript projects.

```bash
npm install @ratel-ai/telemetry-otlp
```

--------------------------------

### init(opts?)

Source: https://docs.ratel.sh/docs/core/api/telemetry-otlp

Initializes the global OpenTelemetry tracer provider with a Ratel OTLP exporter and returns a shutdown handle.

```APIDOC
## init(opts?)

### Description
Wire an OTLP http/protobuf exporter, batch processor, and service name resource over a NodeTracerProvider. It registers the provider globally and returns a shutdown handle.

### Parameters
- **opts** (InitOptions) - Optional - Configuration options for the telemetry initialization.

### Returns
- **TelemetryHandle** - A handle containing a shutdown method to clean up the provider.
```

--------------------------------

### init

Source: https://docs.ratel.sh/llms-full.txt

Initializes the global tracer provider with an OTLP exporter and batch processor.

```APIDOC
## init(*, api_key: str | None = None, endpoint: str | None = None, headers: Mapping[str, str] | None = None, service_name: str | None = None) -> TracerProvider

### Description
Wire an OTLP http/protobuf exporter, batch processor, and service name resource. Registers the provider globally and returns it as a shutdown handle.

### Parameters
- **api_key** (str) - Optional - API key for authentication.
- **endpoint** (str) - Optional - Custom OTLP collector endpoint.
- **headers** (Mapping[str, str]) - Optional - Custom HTTP headers.
- **service_name** (str) - Optional - Name of the service for resource identification.
```

--------------------------------

### Install and Uninstall Skill Hooks CLI

Source: https://docs.ratel.sh/docs/local/managing-skills

Commands to manage the installation and removal of skill hooks with optional scope and confirmation flags.

```bash
ratel-mcp skill install-hook [--scope user|project] [--yes]
ratel-mcp skill uninstall-hook [--scope user|project]
```

--------------------------------

### Install Ratel MCP Plugin

Source: https://docs.ratel.sh/docs/local/quickstart

Commands to add and install the Ratel MCP plugin via Claude or Codex CLI tools.

```bash
claude plugin marketplace add ratel-ai/ratel-mcp
claude plugin install ratel-mcp@ratel
```

```bash
codex plugin marketplace add ratel-ai/ratel-mcp
codex
```

--------------------------------

### size()

Source: https://docs.ratel.sh/llms-full.txt

Get the number of registered skills.

```APIDOC
## size()

### Description
Number of registered skills (distinct ids).

### Returns
- **number** - The count.
```

--------------------------------

### get(skillId)

Source: https://docs.ratel.sh/llms-full.txt

Look up a skill by id.

```APIDOC
## get(skillId)

### Description
Look up a skill by id.

### Parameters
- **skillId** (string) - Required - The id to look up.

### Returns
- **Skill | undefined** - The skill as registered, or undefined for an unknown id.
```

--------------------------------

### Run via CLI

Source: https://docs.ratel.sh/docs/core/framework-integrations

Commands to set the API key and execute the agent using uv.

```bash
export OPENAI_API_KEY=sk-...   # or ANTHROPIC_API_KEY, GEMINI_API_KEY, GROQ_API_KEY
uv run main.py "send an email to alice@example.com saying ship it"
```

--------------------------------

### get()

Source: https://docs.ratel.sh/docs/core/api/sdk-typescript

Look up a tool's searchable metadata.

```APIDOC
## get(toolId)

### Description
Look up a tool's searchable metadata (no executor attached).

### Parameters
#### Path Parameters
- **toolId** (string) - Required - The id to look up.

### Returns
- **Tool | undefined** - The metadata as registered, or undefined for an unknown id.
```

--------------------------------

### Registering Skills and Wiring Search Tools

Source: https://docs.ratel.sh/docs/core/agent-skills

Register skills with dependencies and initialize the search and content retrieval tools.

```python
from ratel_ai import Skill, SkillCatalog, ToolCatalog, get_skill_content_tool, search_capabilities_tool

catalog = ToolCatalog()  # ...register tools...
skills = SkillCatalog()
skills.register(
    Skill(
        id="vercel-deploy",
        name="vercel-deploy",
        description="How to deploy to Vercel: env vars, preview vs production, rollbacks.",
        tags=["deploy", "ship to production"],      # indexed: query terms
        tools=["vercel__deploy", "fs__read_file"],  # dependency edge, not indexed
        metadata={"stacks": ["next", "vercel"]},    # carried context, never matched
        body="## Deploying to Vercel\n1. ...",
    )
)

search = search_capabilities_tool(catalog, skills)  # 2nd argument: skills bucket is live
load = get_skill_content_tool(skills)               # id: "get_skill_content"
```

```typescript
import { SkillCatalog, ToolCatalog, getSkillContentTool, searchCapabilitiesTool } from "@ratel-ai/sdk";

const catalog = new ToolCatalog(); // ...register tools...
const skills = new SkillCatalog();
skills.register({
  id: "vercel-deploy",
  name: "vercel-deploy",
  description: "How to deploy to Vercel: env vars, preview vs production, rollbacks.",
  tags: ["deploy", "ship to production"],     // indexed: query terms
  tools: ["vercel__deploy", "fs__read_file"], // dependency edge, not indexed
  metadata: { stacks: ["next", "vercel"] },   // carried context, never matched
  body: "## Deploying to Vercel\n1. ...",
});

const search = searchCapabilitiesTool(catalog, skills); // 2nd argument: skills bucket is live
const load = getSkillContentTool(skills);               // id: "get_skill_content"
```

--------------------------------

### init(opts?)

Source: https://docs.ratel.sh/llms-full.txt

Wires an OTLP http/protobuf exporter, batch processor, and service name resource over a NodeTracerProvider. It registers the provider globally and returns a shutdown handle.

```APIDOC
## init(opts?)

### Description
Wire an OTLP http/protobuf exporter + batch processor + service.name resource over a NodeTracerProvider, register it as the global provider, and return a shutdown handle.

### Parameters
- **opts** (InitOptions) - Optional - Configuration options for initialization.

### Returns
- **TelemetryHandle** - A handle to manage the shutdown of the telemetry provider.
```

--------------------------------

### get(skillId)

Source: https://docs.ratel.sh/docs/core/api/sdk-typescript

Retrieves a skill object by its ID.

```APIDOC
## get(skillId)

### Description
Looks up a registered skill by its ID.

### Parameters
- **skillId** (string) - Required - The ID to look up.

### Returns
- **Skill | undefined** - The skill object if found, otherwise undefined.
```

--------------------------------

### Register and search tools using BM25

Source: https://docs.ratel.sh/docs/core/api/core-rust

Demonstrates initializing a ToolRegistry, registering tools with JSON schemas, and performing a BM25 search.

```rust
use ratel_ai_core::{Tool, ToolRegistry};

let mut registry = ToolRegistry::new();
registry.register(Tool {
    id: "read_file".into(),
    name: "read_file".into(),
    description: "Read a file from disk".into(),
    input_schema: serde_json::json!({
        "properties": {
            "path": { "type": "string", "description": "absolute path" }
        }
    }),
    output_schema: serde_json::json!({}),
});
registry.register(Tool {
    id: "send_email".into(),
    name: "send_email".into(),
    description: "Send an email to a recipient".into(),
    input_schema: serde_json::json!({}),
    output_schema: serde_json::json!({}),
});

let hits = registry.search("read a file", 5);
assert_eq!(hits[0].tool_id, "read_file");
```

--------------------------------

### get(toolId)

Source: https://docs.ratel.sh/llms-full.txt

Retrieves the searchable metadata for a specific tool by its ID.

```APIDOC
## get(toolId)

### Description
Look up a tool's searchable metadata (no executor attached).

### Parameters
- **toolId** (string) - Required - The id to look up.

### Returns
- **Tool | undefined** - The metadata as registered, or undefined for an unknown id.
```

--------------------------------

### Get Tool Metadata

Source: https://docs.ratel.sh/docs/core/api/sdk-python

Retrieves the metadata-only Tool object for a given identifier.

```python
get(tool_id: str) -> Tool | None
```

--------------------------------

### Import MCP servers

Source: https://docs.ratel.sh/docs/local/quickstart

Run the import wizard in the project directory to migrate existing native MCP servers to Ratel Local.

```bash
ratel-mcp mcp import --agent claude-code
```

```bash
ratel-mcp mcp import --agent codex
```

--------------------------------

### Get Executable Tool

Source: https://docs.ratel.sh/docs/core/api/sdk-python

Retrieves the ExecutableTool object, which includes both metadata and the execution handler.

```python
get_executable(tool_id: str) -> ExecutableTool | None
```

--------------------------------

### Register and Invoke Tools with ToolCatalog

Source: https://docs.ratel.sh/docs/core/sdks/python

Demonstrates the full lifecycle of a tool including registration, semantic search, embedding generation, and invocation.

```python
import asyncio

from ratel_ai import ExecutableTool, ToolCatalog


async def main() -> None:
    catalog = ToolCatalog()                 # ToolCatalog(trace=None, method="bm25")
    catalog.register(                       # metadata + execute
        ExecutableTool(
            id="echo",
            name="echo",
            description="Repeat a message back to the caller.",
            input_schema={"properties": {"text": {"type": "string"}}},
            output_schema={"properties": {"text": {"type": "string"}}},
            execute=lambda args: {"text": args["text"]},
        )
    )
    hits = catalog.search("repeat a message", 5)  # → list[SearchHit]
    assert hits[0].tool_id == "echo"
    catalog.build_embeddings()              # pre-compute semantic/hybrid embeddings
    assert catalog.has("echo")              # → bool
    assert catalog.get("echo") is not None  # → Tool | None
    assert catalog.get_executable("echo") is not None
    print(await catalog.invoke("echo", {"text": "hello"}))
    catalog.drain_trace_events()             # → list[dict]; memory sink only


asyncio.run(main())
```

--------------------------------

### Register a Tool Catalog in Python

Source: https://docs.ratel.sh/llms-full.txt

Create a catalog and register an executable tool, then expose it via capability tools.

```python
from ratel_ai import ToolCatalog, ExecutableTool, search_capabilities_tool, invoke_tool_tool

catalog = ToolCatalog()
catalog.register(
    ExecutableTool(
        id="read_file",
        name="read_file",
        description="Read a file from local disk.",
        input_schema={"properties": {"path": {"type": "string"}}},
        output_schema={"properties": {"contents": {"type": "string"}}},
        execute=lambda args: {"contents": open(args["path"]).read()},
    )
)

# The full catalog stays out of context; the agent reaches it via these two.
search = search_capabilities_tool(catalog)  # id == "search_capabilities"
invoke = invoke_tool_tool(catalog)          # id == "invoke_tool"
```

--------------------------------

### Agent tool execution flow

Source: https://docs.ratel.sh/docs/core/quickstart

Example of the model discovering and invoking a weather tool during a turn.

```text
→ search_capabilities({"query":"weather in Rome"})
→ invoke_tool({"toolId":"get_weather","args":{"city":"Rome"}})

It's 24°C and sunny in Rome.
```

--------------------------------

### Initialize Capability Tools

Source: https://docs.ratel.sh/docs/core/sdks/typescript

Create the search and invoke tools from a catalog to enable agent self-service.

```typescript
const search = searchCapabilitiesTool(catalog); // id: "search_capabilities" (SEARCH_CAPABILITIES_ID)
const invoke = invokeToolTool(catalog);         // id: "invoke_tool"         (INVOKE_TOOL_ID)
```

--------------------------------

### Retrieve Skill Content

Source: https://docs.ratel.sh/docs/core/agent-skills

Example of calling get_skill_content to fetch the body of a specific skill by its ID.

```javascript
get_skill_content({ "skillId": "vercel-deploy" })
```

--------------------------------

### Configure Telemetry in Python

Source: https://docs.ratel.sh/docs/core/telemetry

Initialize the tracer provider with an OTLP endpoint and API key. Ensure the provider is shut down to flush traces on exit.

```python
import os
from ratel_ai import configure_telemetry

provider = configure_telemetry(
    endpoint=os.environ.get("RATEL_URL", "https://cloud.ratel.sh/v1/traces"),
    api_key=os.environ["RATEL_API_KEY"],
)
# ... run your agent: every ratel.* / execute_tool span now exports ...
provider.shutdown()  # flush the exporter on exit (provider.force_flush() mid-run)
```

--------------------------------

### init(options: InitOptions)

Source: https://docs.ratel.sh/llms-full.txt

Initializes the Ratel telemetry provider for a greenfield application.

```APIDOC
## init(options: InitOptions)

### Description
Initializes the Ratel telemetry provider. Returns a TelemetryHandle to manage the lifecycle of the exporter.

### Parameters
- **options** (InitOptions) - Required - Configuration object for the telemetry provider.

### InitOptions Properties
- **apiKey** (string) - Optional - Ratel API key; sent as Authorization: Bearer <apiKey>.
- **endpoint** (string) - Optional - Full OTLP traces URL (incl. /v1/traces). Defaults to RATEL_URL.
- **headers** (Record<string, string>) - Optional - Extra headers merged onto the request.
- **serviceName** (string) - Optional - service.name resource attribute.

### Response
- **TelemetryHandle** - Object containing a shutdown() method to flush and stop the exporter.
```

--------------------------------

### Manage Claude Code statusline

Source: https://docs.ratel.sh/docs/local

Commands to install or uninstall the Ratel statusline into the Claude Code configuration.

```bash
ratel-mcp statusline install     # write the statusLine entry into ~/.claude/settings.json
ratel-mcp statusline uninstall   # remove it, only when Ratel owns it
```

--------------------------------

### InitOptions

Source: https://docs.ratel.sh/docs/core/api/telemetry-typescript

Configuration options for the init() function used to initialize the telemetry exporter.

```APIDOC
## InitOptions

### Description
Configuration object passed to the init() function to set up the telemetry exporter.

### Properties
- **apiKey** (string) - Optional - Ratel API key; sent as Authorization: Bearer <apiKey>.
- **endpoint** (string) - Optional - Full OTLP traces URL (incl. /v1/traces). Defaults to RATEL_URL.
- **headers** (Record<string, string>) - Optional - Extra headers merged onto the request before the apiKey Bearer header.
- **serviceName** (string) - Optional - service.name resource attribute. Defaults to DEFAULT_SERVICE_NAME.
```

--------------------------------

### Skill Hook Settings Configuration

Source: https://docs.ratel.sh/docs/local/managing-skills

The JSON structure appended to the settings file when a skill hook is installed.

```json
{
  "hooks": {
    "UserPromptSubmit": [
      {
        "hooks": [
          { "type": "command", "command": "ratel-mcp skill preload-hook", "timeout": 10 }
        ]
      }
    ]
  }
}
```

--------------------------------

### init()

Source: https://docs.ratel.sh/llms-full.txt

Initializes the telemetry exporter. This function accepts configuration options to define the API key, endpoint, and headers for OTLP trace collection.

```APIDOC
## init(options: InitOptions)

### Description
Initializes the telemetry exporter. This function is provided by @ratel-ai/telemetry-otlp and configures the OTLP exporter settings.

### Parameters
#### InitOptions
- **apiKey** (string) - Optional - Ratel API key; sent as Authorization: Bearer <apiKey>.
- **endpoint** (string) - Optional - Full OTLP traces URL (incl. /v1/traces). Defaults to RATEL_URL.
- **headers** (Record<string, string>) - Optional - Extra headers merged onto the request (before the apiKey Bearer header).
- **serviceName** (string) - Optional - service.name resource attribute. Defaults to DEFAULT_SERVICE_NAME.
```

--------------------------------

### ratelSpanProcessor(opts?)

Source: https://docs.ratel.sh/docs/core/api/telemetry-otlp

Creates a BatchSpanProcessor configured with the Ratel OTLP exporter for use in custom provider setups.

```APIDOC
## ratelSpanProcessor(opts?)

### Description
Creates a BatchSpanProcessor over the Ratel OTLP exporter that forwards spans based on the provided filter. Use this when you want to add Ratel telemetry to an existing provider without global side effects.

### Parameters
- **opts** (RatelSpanProcessorOptions) - Optional - Configuration options including the span filter.

### Returns
- **SpanProcessor** - A configured span processor instance.
```

--------------------------------

### Build search_capabilities tool

Source: https://docs.ratel.sh/docs/core/api/sdk-python

Creates a unified discovery tool for both tools and skills. Recommended for new implementations over search_tools_tool.

```python
search_capabilities_tool(catalog: ToolCatalog, skill_catalog: SkillCatalog | None = None, *, upstream_servers: Sequence[UpstreamServerInfo] | None = None) -> ExecutableTool
```

--------------------------------

### Configure Ratel Local as a Plugin

Source: https://docs.ratel.sh/docs/local

Commands to add and install the Ratel plugin for Claude Code and Codex environments.

```bash
# Claude Code
claude plugin marketplace add ratel-ai/ratel-mcp
claude plugin install ratel-mcp@ratel
# then restart Claude Code or run /reload-plugins

# Codex
codex plugin marketplace add ratel-ai/ratel-mcp
# then install "Ratel Local" from the "Ratel" marketplace inside Codex
```

--------------------------------

### Initialize SkillRegistry

Source: https://docs.ratel.sh/llms-full.txt

Instantiates the metadata-only BM25 index for the skill corpus.

```python
SkillRegistry()
```