Clarify parallel tool-call batching

- document that parallel accepts full reply-shaped tool call objects and ignores planning-only fields
- steer prompt guidance toward one mixed batch for ready independent calls instead of splitting by tool type
- add normalization and prompt regression coverage plus matching DOX notes

helpers/parallel_tools.py.dox.md

@@ -23,6 +23,7 @@
 
 - Helper modules own reusable framework APIs and must preserve public callers unless all callers, tests, and docs are updated together.
 - Wrapped tool-call items must use the same shape as normal tool calls: a tool name plus arguments.
+- Normalization accepts full agent-reply-shaped objects when `tool_name` and `tool_args` are present; non-contract planning fields such as `thoughts` or `headline` are ignored.
 - `call_subordinate` jobs run in isolated child chat contexts tagged with parent-chat metadata; they must not be added to the scheduler task list.
 - Direct tool jobs run in isolated background contexts and are blocked from recursively invoking `parallel`.
 - Parent-visible child log items are created for each wrapped call so the WebUI can inspect concurrent children separately while the wrapper result remains model-history-only.

prompts/agent.system.tool.parallel.md

@@ -1,10 +1,12 @@
 ### parallel
 run independent tool calls concurrently, or await/cancel background parallel jobs.
 
-Use only for independent work. Each `tool_calls` item is a normal tool request object: `{ "tool_name": "...", "tool_args": { ... } }`.
+Use only for independent work. Each `tool_calls` item is a normal tool request object using the same `tool_name` and `tool_args` shape as a top-level reply: `{ "tool_name": "...", "tool_args": { ... } }`.
+Only `tool_name` and `tool_args` are used; if an item is copied from a full reply object, planning fields like `thoughts` or `headline` are ignored.
+Batch all independent calls that are ready now into one `tool_calls` list, even when they use different tools. Do not split by tool type.
 
 Rules:
-- do not use for one simple call, dependent steps, ordered steps, or shared mutable state
+- do not use for one simple call, dependent steps, ordered steps, shared mutable state, or state/tool-availability changes that must happen in the parent context
 - never nest `parallel`
 - `call_subordinate` inside `parallel` starts an isolated child chat under the parent chat, not a scheduler task
 - use `wait: false` only when you will collect results later with `job_ids`
@@ -18,8 +20,8 @@ Start and wait:
   "tool_name": "parallel",
   "tool_args": {
     "tool_calls": [
-      {"tool_name": "call_subordinate", "tool_args": {"message": "Research option A.", "reset": true}},
-      {"tool_name": "call_subordinate", "tool_args": {"message": "Research option B.", "reset": true}}
+      {"tool_name": "call_subordinate", "tool_args": {"message": "Review option A and return key risks.", "reset": true}},
+      {"tool_name": "search_engine", "tool_args": {"query": "official API changelog release notes"}}
     ],
     "wait": true
   }

tests/test_tool_request_normalization.py

@@ -10,6 +10,7 @@ if str(PROJECT_ROOT) not in sys.path:
     sys.path.insert(0, str(PROJECT_ROOT))
 
 from helpers.extract_tools import normalize_tool_request
+from helpers import parallel_tools
 
 
 def test_normalize_tool_request_accepts_canonical_keys() -> None:
@@ -65,3 +66,30 @@ def test_normalize_tool_request_preserves_explicit_action_over_method() -> None:
 def test_normalize_tool_request_rejects_missing_args() -> None:
     with pytest.raises(ValueError, match="tool_args"):
         normalize_tool_request({"tool_name": "response"})
+
+
+def test_normalize_parallel_tool_calls_accepts_full_agent_reply_shape() -> None:
+    calls = parallel_tools.normalize_parallel_tool_calls(
+        [
+            {
+                "thoughts": ["This is independent and ready to run."],
+                "headline": "Search Python release notes",
+                "tool_name": "search_engine",
+                "tool_args": {"query": "latest Python version changelog"},
+            }
+        ]
+    )
+
+    assert calls[0].tool_name == "search_engine"
+    assert calls[0].tool_args == {"query": "latest Python version changelog"}
+
+
+def test_parallel_prompt_encourages_mixed_independent_batches() -> None:
+    prompt = (PROJECT_ROOT / "prompts" / "agent.system.tool.parallel.md").read_text(
+        encoding="utf-8"
+    )
+
+    assert "same `tool_name` and `tool_args` shape as a top-level reply" in prompt
+    assert "planning fields like `thoughts` or `headline` are ignored" in prompt
+    assert "even when they use different tools" in prompt
+    assert "Do not split by tool type" in prompt

tools/parallel.py.dox.md

@@ -21,6 +21,7 @@
 - Tool modules must define `helpers.tool.Tool` subclasses and return `helpers.tool.Response` from `execute(...)`.
 - Wrapped items use the same schema as normal tool calls: a tool name plus arguments.
 - The tool is intended for independent calls only; dependent operations remain sequential.
+- Independent calls should share one batch even when they use different tools; split only for dependencies, ordering, shared mutable state, or parent-context state/tool-availability changes.
 - `action="start"` starts calls and optionally waits according to `wait`.
 - `action="await"` waits for requested job IDs.
 - `action="collect"` returns completed job results without waiting.
@@ -31,6 +32,7 @@
 ## Key Concepts
 
 - `tool_calls`, `calls`, and `items` are accepted aliases for the wrapped call list.
+- Wrapped call items can use the same `tool_name`/`tool_args` shape as top-level agent replies; extra planning fields are ignored by normalization.
 - `job_ids` can be supplied as a string or list when awaiting, collecting, or canceling existing jobs.
 - The response is compact JSON intended for the model to read and continue with.
 - Visible child rows are emitted by `helpers/parallel_tools.py` before each background job starts.