feat(prompts): revamp preset prompts, tool naming, and prompt templating

Rewrite the default/creative/strict presets (classic + agentic) into
structured sections: grounding and cite-by-title guidance, insufficient-
context behavior, current date, respond-in-user-language, scoped mermaid
usage, an untrusted-content guardrail, and a conditional XML-tagged
document context block. A memory directory listing is injected at render
time via the template prefetch mechanism so the model starts oriented
without burning a tool call.

Fixes along the way:
- Agentic preset swap was dead code: _get_prompt_content cached the
  classic preset before create_agent's swap check ran, so agentic and
  research agents always got the classic preset. The swap now happens
  inside _get_prompt_content.
- Jinja autoescape corrupted document content in custom prompts
  (< -> &lt;); prompts are not HTML, autoescape is now off.
- Literal {summaries} leaked into the prompt when no docs were
  retrieved; the placeholder is now stripped.
- Agentic/research prompts referenced tool names from a dropped naming
  scheme (search_internal, reason_think); they now reference the real
  names (search, reason).
- The strict preset told the model to "be very creative and use your
  imagination" right after "never make up information".
- extract_tool_usages recorded intermediate attribute chains as
  bare-tool usages, which meant "run all actions" at prefetch; only
  maximal chains are recorded now.
- Headless runs retrieved docs but never rendered them into the
  prompt; the prompt is now rendered like the streaming path.
- Default tools were unreachable by name in prompt templates
  (prefetch results were keyed by synthetic id only); defaults now
  claim the name key unless an explicit row shadows it.

Tool layer: memory/notes/todo actions are namespaced (memory_view,
note_overwrite, todo_create, ...) with legacy unprefixed names still
accepted via prefix stripping; duplicate action names across tools are
disambiguated with the owning tool's name instead of numeric suffixes;
thin tool descriptions rewritten (brave, duckduckgo, telegram, ntfy,
cryptoprice, read_webpage, internal_search, think).

Docs are now wrapped per chunk in <document index>/<source>/<content>
tags for citation-by-title support.

tests/api/answer/services/test_stream_processor_utils.py

@@ -533,6 +533,34 @@ class TestGetPromptContent:
         # Even with no agent_config, cached value returned
         assert sp._get_prompt_content() == "cached"
 
+    def test_agentic_agent_gets_agentic_preset(self):
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+        sp = StreamProcessor({}, {"sub": "u"})
+        sp.agent_config = {"prompt_id": "default", "agent_type": "agentic"}
+        content = sp._get_prompt_content()
+        assert "`search` tool" in content
+        assert "source.summaries" not in content
+
+    def test_research_agent_gets_agentic_preset(self):
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+        sp = StreamProcessor({}, {"sub": "u"})
+        sp.agent_config = {"prompt_id": "strict", "agent_type": "research"}
+        content = sp._get_prompt_content()
+        assert "`search` tool" in content
+
+    def test_classic_agent_gets_classic_preset(self):
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+        sp = StreamProcessor({}, {"sub": "u"})
+        sp.agent_config = {"prompt_id": "default", "agent_type": "classic"}
+        content = sp._get_prompt_content()
+        assert "source.summaries" in content
+
 
 class TestPreFetchDocs:
     def test_skips_when_no_active_docs(self):
@@ -660,8 +688,67 @@ class TestPreFetchTools:
             d.get("name") == "read_webpage" and d.get("default")
             for d in fetched
         )
-        # Defaults are reachable by synthetic id only — not by name.
         assert default_tool_id("read_webpage") in got
+        # No explicit row of the same name exists, so the default also
+        # claims the name key (what preset templates reference).
+        assert got.get("read_webpage") == {"ok": True}
+
+    def test_explicit_row_keeps_name_key_over_default(self, pg_conn):
+        from application.agents.default_tools import default_tool_id
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+        from application.storage.db.repositories.user_tools import (
+            UserToolsRepository,
+        )
+
+        UserToolsRepository(pg_conn).create(
+            user_id="u-shadow-default", name="read_webpage", status=True
+        )
+        sp = StreamProcessor({}, {"sub": "u-shadow-default"})
+        sp._required_tool_actions = {"read_webpage": {None}}
+
+        def _fake_fetch(tool_doc, required_actions):
+            return {"is_default": bool(tool_doc.get("default"))}
+
+        with _patch_db(pg_conn), patch(
+            "application.api.answer.services.stream_processor.settings.ENABLE_TOOL_PREFETCH",
+            True,
+        ), patch.object(sp, "_fetch_tool_data", _fake_fetch):
+            got = sp.pre_fetch_tools()
+        assert got is not None
+        # The explicit row owns the name key; the default stays reachable
+        # by its synthetic id.
+        assert got["read_webpage"] == {"is_default": False}
+        assert got[default_tool_id("read_webpage")] == {"is_default": True}
+
+    def test_fetch_tool_data_executes_referenced_memory_view(self, pg_conn):
+        from unittest.mock import MagicMock
+
+        from application.agents.default_tools import synthesize_default_tool
+        from application.api.answer.services.stream_processor import (
+            StreamProcessor,
+        )
+
+        sp = StreamProcessor({}, {"sub": "u-mem-prefetch"})
+        tool_doc = synthesize_default_tool("memory")
+
+        mock_tool = MagicMock()
+        mock_tool.get_actions_metadata.return_value = tool_doc["actions"]
+        mock_tool.execute_action.return_value = "Directory: /\n(empty)"
+        mock_manager = MagicMock()
+        mock_manager.load_tool.return_value = mock_tool
+
+        with patch(
+            "application.agents.tools.tool_manager.ToolManager",
+            return_value=mock_manager,
+        ):
+            got = sp._fetch_tool_data(tool_doc, {"memory_view"})
+
+        # Only the referenced action ran, with no path kwarg — the tool's
+        # own "/" default applies.
+        assert got == {"memory_view": "Directory: /\n(empty)"}
+        mock_tool.execute_action.assert_called_once_with("memory_view")
 
     def test_agent_bound_invocation_omits_default_tool_prefetch(self, pg_conn):
         from application.api.answer.services.stream_processor import (