feat: add agent integration — Python SDK, MCP server, LLM tool-use schema

- sit/sdk.py: Python SDK with Sit class wrapping all CLI commands as direct API calls
- sit/mcp_server.py: MCP stdio server exposing 7 tools (info/validate/test/diff/pr_summary/report/doctor)
- sit/tool_use.py: LLM tool-use JSON Schema for OpenAI and Anthropic function calling
- pyproject.toml: add [mcp] optional deps group and sit-mcp-server entry point
- README.md: add Agent Integration section documenting SDK, MCP, and tool-use usage

Author: SitHub Maintainer

README.md

@@ -120,6 +120,60 @@ sit test "$SIT_PACKAGE_DIR" --run
 sit ci-summary "$SIT_PACKAGE_DIR" --compare origin/main..HEAD >> "$GITHUB_STEP_SUMMARY"
 ```
 
+## Agent Integration
+
+`sit` can be called by AI agents in three ways:
+
+### Python SDK
+
+```python
+from sit.sdk import Sit
+
+s = Sit("./my-skill-package")
+info = s.info()          # sit.info.v1 contract
+test = s.test()          # sit.test.v1 contract
+diff = s.diff("./old")   # sit.diff.v1 contract
+pr = s.pr_summary("./old")  # sit.pr_summary.v1 contract
+report = s.report(compare="./old")  # sit.report.v1 contract
+```
+
+### MCP Server
+
+Install with MCP support and run the stdio server:
+
+```bash
+pip install 'sit-toolkit[mcp]'
+sit-mcp-server
+```
+
+Or configure in your MCP client (e.g., Claude Desktop):
+
+```json
+{
+  "mcpServers": {
+    "sit": {
+      "command": "sit-mcp-server"
+    }
+  }
+}
+```
+
+Exposes 7 tools: `sit_info`, `sit_validate`, `sit_test`, `sit_diff`, `sit_pr_summary`, `sit_report`, `sit_doctor`.
+
+### LLM Tool-Use Schema
+
+```python
+from sit.tool_use import get_tools_openai, get_tools_anthropic
+
+# For OpenAI
+tools = get_tools_openai()
+response = client.chat_completion(messages=..., tools=tools)
+
+# For Anthropic Claude
+tools = get_tools_anthropic()
+response = client.messages.create(messages=..., tools=tools)
+```
+
 ## Research
 
 `sit` has been validated through multi-agent experiments on AI Skill packaging workflows. See:

pyproject.toml

@@ -39,6 +39,9 @@ dev = [
 binary = [
   "pyinstaller>=6.0",
 ]
+mcp = [
+  "mcp>=1.0",
+]
 
 [project.urls]
 Homepage = "https://github.com/OpenRaiser/SitHub"
@@ -48,6 +51,7 @@ Issues = "https://github.com/OpenRaiser/SitHub/issues"
 
 [project.scripts]
 sit = "sit.cli:main"
+sit-mcp-server = "sit.mcp_server:main"
 
 [tool.setuptools]
 packages = ["sit"]

sit/__init__.py

@@ -1,3 +1,3 @@
 """Skill Iteration Toolkit."""
 
-__version__ = "0.19.0"
+__version__ = "0.20.0"

sit/mcp_server.py

@@ -0,0 +1,273 @@
+"""MCP (Model Context Protocol) server for sit.
+
+Exposes sit commands as MCP tools so LLM agents can call them natively
+via the MCP protocol.  Uses the Python SDK under the hood — no subprocess.
+
+Usage::
+
+    # Run as stdio MCP server (default transport):
+    python -m sit.mcp_server
+
+    # Or import and customise:
+    from sit.mcp_server import create_server
+    server = create_server()
+    server.run()
+"""
+
+from __future__ import annotations
+
+import json
+from pathlib import Path
+from typing import Any
+
+try:
+    from mcp.server import Server
+    from mcp.server.stdio import stdio_server
+    from mcp.types import TextContent, Tool
+except ImportError:
+    raise ImportError(
+        "MCP support requires the 'mcp' package. "
+        "Install it with: pip install 'sit-toolkit[mcp]' or pip install mcp"
+    )
+
+from .sdk import Sit
+
+# ---------------------------------------------------------------------------
+# Tool definitions
+# ---------------------------------------------------------------------------
+
+TOOLS: list[Tool] = [
+    Tool(
+        name="sit_info",
+        description=(
+            "Get package info for a sit skill package. "
+            "Returns the sit.info.v1 contract: name, version, git state, "
+            "file listing, validation status, golden test results, reports."
+        ),
+        inputSchema={
+            "type": "object",
+            "properties": {
+                "package_path": {
+                    "type": "string",
+                    "description": "Path to the skill package directory or skill.yaml file.",
+                },
+            },
+            "required": ["package_path"],
+        },
+    ),
+    Tool(
+        name="sit_validate",
+        description=(
+            "Validate a sit skill package structure. "
+            "Checks skill.yaml fields, prompt/schema/test paths, schema validity, "
+            "JSONL format. Returns pass/fail with messages."
+        ),
+        inputSchema={
+            "type": "object",
+            "properties": {
+                "package_path": {
+                    "type": "string",
+                    "description": "Path to the skill package directory or skill.yaml file.",
+                },
+            },
+            "required": ["package_path"],
+        },
+    ),
+    Tool(
+        name="sit_test",
+        description=(
+            "Run validation + golden schema tests on a sit skill package. "
+            "Returns the sit.test.v1 contract with validation and test results."
+        ),
+        inputSchema={
+            "type": "object",
+            "properties": {
+                "package_path": {
+                    "type": "string",
+                    "description": "Path to the skill package directory or skill.yaml file.",
+                },
+                "run_actual": {
+                    "type": "boolean",
+                    "description": "If true, run the actual test runner (commands.run_case) instead of static validation only.",
+                    "default": False,
+                },
+                "runner": {
+                    "type": "string",
+                    "description": "Override the test runner command (default: from skill.yaml commands.run_case).",
+                },
+                "timeout": {
+                    "type": "integer",
+                    "description": "Runner timeout in seconds (default: 30).",
+                    "default": 30,
+                },
+            },
+            "required": ["package_path"],
+        },
+    ),
+    Tool(
+        name="sit_diff",
+        description=(
+            "Semantic diff between two sit skill packages. "
+            "Returns the sit.diff.v1 contract with change events, risk assessment, "
+            "and suggested version bump."
+        ),
+        inputSchema={
+            "type": "object",
+            "properties": {
+                "old_path": {
+                    "type": "string",
+                    "description": "Path to the baseline/old skill package.",
+                },
+                "new_path": {
+                    "type": "string",
+                    "description": "Path to the current/new skill package.",
+                },
+                "include_text_diffs": {
+                    "type": "boolean",
+                    "description": "Include full unified diff lines in text_diffs (default: false).",
+                    "default": False,
+                },
+            },
+            "required": ["old_path", "new_path"],
+        },
+    ),
+    Tool(
+        name="sit_pr_summary",
+        description=(
+            "Generate a PR summary for a skill package change. "
+            "Returns the sit.pr_summary.v1 contract with baseline/current refs, "
+            "validation, tests, risk, and semantic diff."
+        ),
+        inputSchema={
+            "type": "object",
+            "properties": {
+                "baseline_path": {
+                    "type": "string",
+                    "description": "Path to the baseline skill package.",
+                },
+                "current_path": {
+                    "type": "string",
+                    "description": "Path to the current skill package.",
+                },
+            },
+            "required": ["baseline_path", "current_path"],
+        },
+    ),
+    Tool(
+        name="sit_report",
+        description=(
+            "Build a full sit report for a skill package. "
+            "Returns the sit.report.v1 contract with validation, golden tests, "
+            "optional diff, and reproducibility commands."
+        ),
+        inputSchema={
+            "type": "object",
+            "properties": {
+                "package_path": {
+                    "type": "string",
+                    "description": "Path to the skill package directory or skill.yaml file.",
+                },
+                "compare_path": {
+                    "type": "string",
+                    "description": "Optional path to a baseline package for diff comparison.",
+                },
+            },
+            "required": ["package_path"],
+        },
+    ),
+    Tool(
+        name="sit_doctor",
+        description=(
+            "Run environment diagnostics for a sit skill package. "
+            "Checks git, GitHub remote, manifest, validation, and golden tests."
+        ),
+        inputSchema={
+            "type": "object",
+            "properties": {
+                "package_path": {
+                    "type": "string",
+                    "description": "Path to the skill package directory or skill.yaml file.",
+                },
+            },
+            "required": ["package_path"],
+        },
+    ),
+]
+
+
+# ---------------------------------------------------------------------------
+# Handler
+# ---------------------------------------------------------------------------
+
+def _handle_tool(name: str, arguments: dict[str, Any]) -> str:
+    """Dispatch an MCP tool call to the SDK and return JSON string."""
+    try:
+        if name == "sit_info":
+            result = Sit(arguments["package_path"]).info()
+        elif name == "sit_validate":
+            result = Sit(arguments["package_path"]).validate()
+        elif name == "sit_test":
+            result = Sit(arguments["package_path"]).test(
+                run_actual=arguments.get("run_actual", False),
+                runner=arguments.get("runner"),
+                timeout=arguments.get("timeout", 30),
+            )
+        elif name == "sit_diff":
+            result = Sit(arguments["new_path"]).diff(
+                arguments["old_path"],
+                include_text_diffs=arguments.get("include_text_diffs", False),
+            )
+        elif name == "sit_pr_summary":
+            result = Sit(arguments["current_path"]).pr_summary(arguments["baseline_path"])
+        elif name == "sit_report":
+            result = Sit(arguments["package_path"]).report(
+                compare=arguments.get("compare_path"),
+            )
+        elif name == "sit_doctor":
+            result = Sit(arguments["package_path"]).doctor()
+        else:
+            return json.dumps({"error": f"Unknown tool: {name}"})
+        return json.dumps(result, ensure_ascii=False, indent=2)
+    except Exception as exc:
+        return json.dumps({"error": f"{type(exc).__name__}: {exc}"})
+
+
+# ---------------------------------------------------------------------------
+# Server factory
+# ---------------------------------------------------------------------------
+
+def create_server() -> Server:
+    """Create and configure an MCP Server instance with sit tools."""
+    server = Server("sit-mcp-server")
+
+    @server.list_tools()
+    async def list_tools() -> list[Tool]:
+        return TOOLS
+
+    @server.call_tool()
+    async def call_tool(name: str, arguments: dict[str, Any]) -> list[TextContent]:
+        result = _handle_tool(name, arguments)
+        return [TextContent(type="text", text=result)]
+
+    return server
+
+
+# ---------------------------------------------------------------------------
+# Entry point
+# ---------------------------------------------------------------------------
+
+async def _run_stdio() -> None:
+    server = create_server()
+    async with stdio_server() as (read_stream, write_stream):
+        await server.run(read_stream, write_stream, server.create_initialization_options())
+
+
+def main() -> None:
+    """Run the sit MCP server over stdio transport."""
+    import asyncio
+
+    asyncio.run(_run_stdio())
+
+
+if __name__ == "__main__":
+    main()