fix(tools): page through MCP tools/list nextCursor

raven/agent/tools/mcp.py

@@ -13,6 +13,8 @@
 from raven.sandbox import SandboxInitError
 
 if TYPE_CHECKING:
+    from mcp import ClientSession
+
     from raven.sandbox import SandboxExecutor
 
 
@@ -75,6 +77,26 @@ async def execute(self, **kwargs: Any) -> str | ToolResult:
         return text or "(no output)"
 
 
+async def _collect_tools(session: "ClientSession") -> list:
+    """Page through tools/list until the server stops returning a nextCursor.
+
+    A server may paginate tool discovery when it exposes more tools than its
+    page size. Track seen cursors so a misbehaving server that re-issues the
+    same cursor cannot trap discovery in an infinite loop.
+    """
+    all_tools = []
+    seen_cursors = {None}
+    cursor = None
+    while True:
+        page = await session.list_tools(cursor)
+        all_tools.extend(page.tools)
+        cursor = page.nextCursor
+        if not cursor or cursor in seen_cursors:
+            break
+        seen_cursors.add(cursor)
+    return all_tools
+
+
 async def connect_mcp_servers(
     mcp_servers: dict,
     registry: ToolRegistry,
@@ -158,13 +180,14 @@ def httpx_client_factory(
             session = await stack.enter_async_context(ClientSession(read, write))
             await session.initialize()
 
-            tools = await session.list_tools()
-            for tool_def in tools.tools:
+            all_tools = await _collect_tools(session)
+
+            for tool_def in all_tools:
                 wrapper = MCPToolWrapper(session, name, tool_def, tool_timeout=cfg.tool_timeout)
                 registry.register(wrapper)
                 logger.debug("MCP: registered tool '{}' from server '{}'", wrapper.name, name)
 
-            logger.info("MCP server '{}': connected, {} tools registered", name, len(tools.tools))
+            logger.info("MCP server '{}': connected, {} tools registered", name, len(all_tools))
         except (Exception, BaseExceptionGroup) as e:
             # BaseExceptionGroup is raised by anyio task groups (e.g. streamableHttp cancel
             # scope failures) and is not a subclass of Exception in Python 3.11+.

tests/test_mcp_tools_pagination.py

@@ -0,0 +1,64 @@
+"""Pagination tests for MCP tools/list discovery (issue #301)."""
+
+from __future__ import annotations
+
+from types import SimpleNamespace
+
+from raven.agent.tools.mcp import _collect_tools
+
+
+def _page(tool_names: list[str], next_cursor: str | None = None) -> SimpleNamespace:
+    return SimpleNamespace(
+        tools=[SimpleNamespace(name=n) for n in tool_names],
+        nextCursor=next_cursor,
+    )
+
+
+class _FakeSession:
+    """Replays a scripted list_tools response keyed by the cursor argument."""
+
+    def __init__(self, pages: dict):
+        self._pages = pages
+        self.calls: list = []
+
+    async def list_tools(self, cursor: str | None = None) -> SimpleNamespace:
+        self.calls.append(cursor)
+        return self._pages[cursor]
+
+
+async def test_single_page_returns_tools_without_follow_up() -> None:
+    session = _FakeSession({None: _page(["a", "b"])})
+
+    tools = await _collect_tools(session)
+
+    assert [t.name for t in tools] == ["a", "b"]
+    assert session.calls == [None]
+
+
+async def test_multi_page_follows_cursor_chain() -> None:
+    session = _FakeSession(
+        {
+            None: _page(["a"], "c1"),
+            "c1": _page(["b"], "c2"),
+            "c2": _page(["c"]),
+        }
+    )
+
+    tools = await _collect_tools(session)
+
+    assert [t.name for t in tools] == ["a", "b", "c"]
+    assert session.calls == [None, "c1", "c2"]
+
+
+async def test_repeated_cursor_stops_instead_of_looping() -> None:
+    session = _FakeSession(
+        {
+            None: _page(["a"], "c1"),
+            "c1": _page(["b"], "c1"),
+        }
+    )
+
+    tools = await _collect_tools(session)
+
+    assert [t.name for t in tools] == ["a", "b"]
+    assert session.calls == [None, "c1"]