Skip to content

Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.

fix(learn/claude): don't abort the whole scan on a null message line (#2299)

Abhay Singheed80dd4baa6
diff --git a/headroom/learn/plugins/claude.py b/headroom/learn/plugins/claude.py
index 185ba32de..399cacfba 100644
--- a/headroom/learn/plugins/claude.py
+++ b/headroom/learn/plugins/claude.py
@@ -208,7 +208,12 @@ class ClaudeCodePlugin(LearnPlugin, ConversationScanner):
if line_type == "assistant":
self._extract_tool_uses(d, tool_uses)
- usage = d.get("message", {}).get("usage", {})
+ # `get("message", {})` returns None for an explicit
+ # {"message": null} line (the default only applies to a
+ # missing key); `.get` on None then raises AttributeError,
+ # which the OSError/UnicodeDecodeError guard does not catch
+ # — so one malformed line crashed the whole learn run.
+ usage = (d.get("message") or {}).get("usage", {})
total_input_tokens += usage.get("input_tokens", 0)
total_input_tokens += usage.get("cache_read_input_tokens", 0)
total_input_tokens += usage.get("cache_creation_input_tokens", 0)
@@ -237,7 +242,7 @@ class ClaudeCodePlugin(LearnPlugin, ConversationScanner):
def _extract_tool_uses(self, d: dict, tool_uses: dict[str, tuple[str, dict]]) -> None:
"""Extract tool_use blocks from an assistant message."""
- msg = d.get("message", {})
+ msg = d.get("message") or {}
content = msg.get("content", [])
if not isinstance(content, list):
return
@@ -261,7 +266,7 @@ class ClaudeCodePlugin(LearnPlugin, ConversationScanner):
timestamp: str | None = None,
) -> None:
"""Extract tool_result blocks from a user message and match to tool_uses."""
- msg = d.get("message", {})
+ msg = d.get("message") or {}
content = msg.get("content", [])
if not isinstance(content, list):
return
@@ -327,7 +332,7 @@ class ClaudeCodePlugin(LearnPlugin, ConversationScanner):
timestamp: str | None = None,
) -> None:
"""Extract user text messages and interruptions from a user line."""
- msg = d.get("message", {})
+ msg = d.get("message") or {}
content = msg.get("content", "")
if isinstance(content, str) and content.strip():
diff --git a/tests/test_learn/test_subagent_scanning.py b/tests/test_learn/test_subagent_scanning.py
index f533a0859..8839cf2aa 100644
--- a/tests/test_learn/test_subagent_scanning.py
+++ b/tests/test_learn/test_subagent_scanning.py
@@ -42,6 +42,46 @@ def _write_session(path: Path, out: str = "x" * 400) -> None:
path.write_text("\n".join(json.dumps(line) for line in lines))
+def test_scan_survives_null_message_line(tmp_path: Path) -> None:
+ # A single line with an explicit {"message": null} must not crash the scan
+ # (the per-file guard only catches OSError/UnicodeDecodeError, so an
+ # AttributeError here would abort the whole `learn` run). The valid tool_use
+ # pair around it must still be parsed.
+ path = tmp_path / "main-uuid.jsonl"
+ lines = [
+ {"type": "assistant", "message": None},
+ {
+ "type": "assistant",
+ "message": {
+ "usage": {"input_tokens": 100, "output_tokens": 10},
+ "content": [
+ {
+ "type": "tool_use",
+ "id": "u1",
+ "name": "Read",
+ "input": {"file_path": "/a.py"},
+ }
+ ],
+ },
+ },
+ {"type": "user", "message": None},
+ {
+ "type": "user",
+ "message": {
+ "content": [{"type": "tool_result", "tool_use_id": "u1", "content": "x" * 400}]
+ },
+ },
+ ]
+ path.write_text("\n".join(json.dumps(line) for line in lines))
+
+ plugin = ClaudeCodePlugin()
+ session = plugin._scan_session(path)
+
+ assert session is not None
+ assert len(session.tool_calls) == 1
+ assert session.total_input_tokens == 100
+
+
def test_scan_project_discovers_subagent_and_workflow_transcripts(tmp_path: Path) -> None:
_write_session(tmp_path / "main-uuid.jsonl")
_write_session(tmp_path / "main-uuid" / "subagents" / "agent-1.jsonl")