Ajouter une compétence

---

title: Add Skill api: POST /api/v1/skills

Add a skill to the knowledge base. Skills define capabilities that agents can invoke. OpenViking automatically detects and converts MCP tool definitions to skill format.

<Note> Skills are vectorized and indexed for semantic retrieval. Use `wait=true` or call `/api/v1/system/wait` to ensure processing completes. </Note>

Authentication

Requires API key authentication via X-API-Key header.

Request Body

<ParamField body="data" type="object | string" required> Skill data in one of four supported formats:

1. Skill format (dict with name, description, content)
2. MCP Tool format (dict with inputSchema - auto-converted)
3. String (SKILL.md content with YAML frontmatter)
4. Path (file or directory path) </ParamField>

<ParamField body="wait" type="boolean" default="false"> Wait for vectorization to complete before returning </ParamField> <ParamField body="timeout" type="number" default="null"> Timeout in seconds when `wait=true` </ParamField>

Skill Data Formats

Format 1: Skill Dict

{
  "name": "search-web",
  "description": "Search the web for current information",
  "content": "# search-web\n\nSearch the web for current information.\n\n## Parameters\n- **query** (string, required): Search query\n- **limit** (integer, optional): Max results, default 10",
  "allowed_tools": ["HttpClient", "JsonParser"],
  "tags": ["web", "search"]
}

Format 2: MCP Tool (Auto-Converted)

{
  "name": "calculator",
  "description": "Perform mathematical calculations",
  "inputSchema": {
    "type": "object",
    "properties": {
      "expression": {
        "type": "string",
        "description": "Mathematical expression to evaluate"
      }
    },
    "required": ["expression"]
  }
}

<Info> OpenViking detects MCP format by the presence of `inputSchema` and automatically converts it to skill format with generated markdown documentation. </Info>

Format 3: SKILL.md String

---
name: code-runner
description: Execute code in various languages
allowed-tools:
  - FileSystem
  - ProcessRunner
tags:
  - code
  - execution
---

# code-runner

Execute code in various languages safely.

## Parameters
- **language** (string, required): Programming language
- **code** (string, required): Code to execute
- **timeout** (integer, optional): Timeout in seconds

Format 4: File Path

{
  "data": "./skills/search-web/SKILL.md"
}

Or for directories with auxiliary files:

{
  "data": "./skills/code-runner/"
}

Response

<ResponseField name="status" type="string"> Response status (`ok` or `error`) </ResponseField> <ResponseField name="result" type="object"> Skill creation result <Expandable title="Result Object"> <ResponseField name="status" type="string"> Creation status (`success`) </ResponseField>

<ResponseField name="uri" type="string">
  URI where skill was stored (e.g., `viking://agent/skills/search-web/`)
</ResponseField>

<ResponseField name="name" type="string">
  Skill name in kebab-case
</ResponseField>

<ResponseField name="auxiliary_files" type="number">
  Number of auxiliary files included (when adding from directory)
</ResponseField>

</Expandable> </ResponseField> <ResponseField name="time" type="number"> Request processing time in seconds </ResponseField>

Examples

<CodeGroup>

curl -X POST http://localhost:1933/api/v1/skills \
  -H "Content-Type: application/json" \
  -H "X-API-Key: your-key" \
  -d '{
    "data": {
      "name": "search-web",
      "description": "Search the web for current information",
      "content": "# search-web\n\nSearch the web for current information.\n\n## Parameters\n- **query** (string, required): Search query\n- **limit** (integer, optional): Max results, default 10"
    }
  }'

curl -X POST http://localhost:1933/api/v1/skills \
  -H "Content-Type: application/json" \
  -H "X-API-Key: your-key" \
  -d '{
    "data": {
      "name": "calculator",
      "description": "Perform mathematical calculations",
      "inputSchema": {
        "type": "object",
        "properties": {
          "expression": {
            "type": "string",
            "description": "Mathematical expression to evaluate"
          }
        },
        "required": ["expression"]
      }
    }
  }'

# Add from skill dict
skill = {
    "name": "search-web",
    "description": "Search the web for current information",
    "content": """
# search-web

Search the web for current information.

## Parameters
- **query** (string, required): Search query
- **limit** (integer, optional): Max results, default 10
"""
}
result = client.add_skill(skill)
print(f"Added: {result['uri']}")

# Add from MCP tool (auto-converted)
mcp_tool = {
    "name": "calculator",
    "description": "Perform mathematical calculations",
    "inputSchema": {
        "type": "object",
        "properties": {
            "expression": {
                "type": "string",
                "description": "Mathematical expression to evaluate"
            }
        },
        "required": ["expression"]
    }
}
result = client.add_skill(mcp_tool)
print(f"Added: {result['uri']}")

# Add from file
result = client.add_skill("./skills/search-web/SKILL.md")
print(f"Added: {result['uri']}")

# Add from directory (includes auxiliary files)
result = client.add_skill("./skills/code-runner/")
print(f"Added: {result['uri']}")
print(f"Auxiliary files: {result['auxiliary_files']}")

# Add from file
openviking add-skill ./skills/search-web/SKILL.md

# Add from directory and wait for processing
openviking add-skill ./skills/code-runner/ --wait

</CodeGroup> <ResponseExample>

{
  "status": "ok",
  "result": {
    "status": "success",
    "uri": "viking://agent/skills/search-web/",
    "name": "search-web",
    "auxiliary_files": 0
  },
  "time": 0.15
}

{
  "status": "ok",
  "result": {
    "status": "success",
    "uri": "viking://agent/skills/code-runner/",
    "name": "code-runner",
    "auxiliary_files": 3
  },
  "time": 0.25
}

</ResponseExample>

MCP Tool Conversion

When you provide a dict with inputSchema, OpenViking automatically converts it to skill format:

Input (MCP):

{
  "name": "search_web",
  "description": "Search the web",
  "inputSchema": {
    "type": "object",
    "properties": {
      "query": {"type": "string", "description": "Search query"},
      "limit": {"type": "integer", "description": "Max results"}
    },
    "required": ["query"]
  }
}

Output (Skill):

---
name: search-web
description: Search the web
---

# search-web

Search the web

## Parameters

- **query** (string) (required): Search query
- **limit** (integer) (optional): Max results

## Usage

This tool wraps the MCP tool `search-web`. Call this when the user needs functionality matching the description above.

Best Practices

<AccordionGroup> <Accordion title="Use Clear Descriptions"> ```python # Good - specific and actionable skill = { "name": "search-web", "description": "Search the web for current information using Google", ... }

# Less helpful - too vague
skill = {
    "name": "search",
    "description": "Search",
    ...
}
```

</Accordion> <Accordion title="Use Kebab-Case for Names"> - `search-web` ✅ - `searchWeb` ❌ - `search_web` ❌ </Accordion> <Accordion title="Include Comprehensive Content"> - Clear parameter descriptions with types - When to use the skill - Concrete examples - Edge cases and limitations </Accordion> </AccordionGroup>

Related Endpoints

• List Skills - View all skills
• Search Skills - Semantic search
• Remove Skill - Delete a skill
• Wait for Processing - Ensure vectorization completes