mirror of
https://github.com/headroomlabs-ai/headroom.git
synced 2026-11-22 18:41:07 +08:00
## Description When the model calls `headroom_retrieve` in parallel with client tools, `CCRResponseHandler.handle_response` hands the whole turn back for the client to resolve (#839). The client has no `headroom_retrieve`, so it can never resolve that call: Claude Code answers `Error: No such tool available: headroom_retrieve`, and the model never gets the content it asked for. Parallel calls are the normal shape for agentic clients. A Claude Code subagent that greps and reads several files in one turn hits this regularly, and users report the error as "happens sometimes, usually with subagents". Measured on one machine's Claude Code transcripts (`~/.claude/projects`): 501 leaked `headroom_retrieve` calls since 2026-09-21. 484 of them were mixed turns (retrieve plus at least one client tool in the same assistant message), and 435 of those were in subagents. The other 17 were lone retrieves on the #3071 streaming path, which this PR does not touch. The client tools in a mixed turn have not run yet. This PR drops them from the turn, serves the retrieval in the continuation exactly as for a lone call, and lets the model decide again there, with the retrieved content in hand. The continuation is what the client receives, whatever it does: re-issue the client calls, change them, or withdraw them because the retrieved content says not to. The client never sees `headroom_retrieve` and never runs a call the model withdrew. Only when the continuation request fails does the client get the model's own mixed turn back (the #839 behaviour), since the model made no newer decision; a continuation that succeeded is kept even if a later retrieval round fails. A `headroom_retrieve` without a valid hash is never treated as a client call: such a turn is handed back unchanged. This applies to Anthropic and OpenAI chat only. In the Responses API a `function_call` can be the item a `reasoning` item requires next, and a Gemini `functionCall` part can carry the turn's thought signature, so dropping a sibling there could make the continuation invalid. Those providers keep the #839 pass-through unchanged. ## Type of Change - [x] Bug fix (non-breaking change that fixes an issue) - [ ] New feature (non-breaking change that adds functionality) - [ ] Breaking change (fix or feature that would cause existing functionality to change) - [ ] Documentation update - [ ] Performance improvement - [ ] Code refactoring (no functional changes) ## Changes Made - `ccr/tool_calls.py`: new `drop_tool_calls(response, provider, calls)`. It returns a shallow copy of the response without the given tool-call objects, matched by identity. It handles Anthropic and OpenAI chat and returns the response unchanged for any other provider. - `ccr/response_handler.py`: on a mixed turn, `handle_response` trims the client calls and continues the normal retrieval loop. A successful continuation is returned as is. If the continuation request for a mixed round fails, it returns the original mixed turn instead of the trimmed one. When the provider is not trimmable, or one of the extra calls is a `headroom_retrieve` without a valid hash, it keeps the old skip-and-return. The model's own response object is not mutated. - Tests: the #839 handler test now asserts the new behaviour for Anthropic and OpenAI chat: one continuation carrying only the retrieve pair, and the re-issuing continuation returned. New handler tests return the continuation when it withdraws the client call in text, changes its arguments, or re-issues it unchanged; return the original turn when the continuation request fails; and keep a successful continuation when a later retrieval round fails. A malformed `headroom_retrieve` keeps the turn unchanged with no continuation. The Responses pass-through is pinned as unchanged. On the streaming SSE path, a re-issued client tool streams without `headroom_retrieve`, a continuation that withdraws the client tool streams the continuation, and a failed continuation streams both original blocks as a 200. ## Testing - [x] Unit tests pass (`pytest`) - [x] Linting passes (`ruff check .`) - [x] Type checking passes (`mypy headroom`) - [x] New tests added for new functionality - [x] Manual testing performed ### Test Output ```text $ uv run --frozen --extra dev pytest tests/ -q -k "ccr or CCR or retrieve or mixed" 1168 passed, 24 skipped, 15064 deselected, 3 warnings in 178.17s $ uv run --frozen --extra dev pytest <every test file naming headroom_retrieve, CCR_TOOL_NAME or ccr_response_handler> # at03e8d8761101 passed, 63 skipped, 3 warnings in 86.75s # against main (tests kept): the serve tests fail FAILED tests/test_ccr_response_handler.py::TestCCRResponseHandling::test_handle_response_mixed_tools_serves_retrieval FAILED tests/test_ccr_response_handler.py::TestCCRResponseHandling::test_handle_response_mixed_tools_serves_retrieval_openai FAILED tests/test_proxy/test_anthropic_streaming_ccr_retrieve.py::test_mixed_ccr_and_client_tool_serves_retrieval_then_streams_reissued_tool # against95dc4975(handed the original turn back unless every client call was re-issued unchanged): the continuation tests fail FAILED tests/test_ccr_response_handler.py::TestCCRResponseHandling::test_handle_response_mixed_tools_returns_the_continuation[withdrawn] FAILED tests/test_ccr_response_handler.py::TestCCRResponseHandling::test_handle_response_mixed_tools_returns_the_continuation[changed] FAILED tests/test_ccr_response_handler.py::TestCCRResponseHandling::test_handle_response_mixed_tools_later_failure_keeps_the_continuation FAILED tests/test_proxy/test_anthropic_streaming_ccr_retrieve.py::test_mixed_turn_streams_the_continuation_or_the_turn_on_failure[withdrawn] # against the first commit of this PR (trimmed turn returned on a failed continuation, malformed retrieve trimmed) FAILED tests/test_ccr_response_handler.py::TestCCRResponseHandling::test_handle_response_mixed_tools_failed_continuation_returns_turn FAILED tests/test_ccr_response_handler.py::TestCCRResponseHandling::test_handle_response_malformed_ccr_call_keeps_turn $ uv run --frozen --extra dev ruff check <changed files> && ruff format --check <changed files> All checks passed! $ uv run --frozen --extra dev mypy headroom/ccr/tool_calls.py headroom/ccr/response_handler.py Success: no issues found in 2 source files ``` ## Real Behavior Proof - Environment: macOS, the released `headroom-ai==0.39.0` wheel (byte-identical `response_handler.py` and `tool_calls.py` to main3c418f6f) with this logic applied to the installed `handle_response`, the way Headroom Desktop ships it. - Exact command / steps: ran `CCRResponseHandler().handle_response` on a stored hash with an Anthropic turn `[tool_use headroom_retrieve, tool_use Read]` and a fake continuation that re-issues the Read, then one that answers in text, then the same turn shaped as Responses `function_call` items. - Observed result: re-issued: one continuation carrying only the retrieve `tool_use` and a `tool_result` with the original content, and the client gets the re-issued `[Read]`. Answered in text: the client gets that text, no calls. The reviewer's loopback-HTTP reproducer (#issuecomment-6062651317) on03e8d876, with its fallback assertion inverted: a cancelling text continuation returns no call, a changed `write_file(new.txt)` returns that call, an unchanged re-issue returns it, and the original `write_file(old.txt)` is never handed back. Without the change, no continuation and the client gets `[headroom_retrieve, Read]`. The Responses turn comes back unchanged in every case. - Not tested: a live Anthropic call where the trimmed assistant turn also carries a signed `thinking` block. Only `tool_use` blocks are removed and thinking blocks are kept unmodified, which is the shape the lone-retrieve continuation already sends. ## Runtime Rollout Safety - Rollout-managed feature(s): none; CCR response handling (`ccr_handle_responses`) is existing behaviour. - Minimum rollout channel: n/a - Stable/default behavior changed: yes; a mixed retrieve plus client-tool turn on Anthropic or OpenAI chat is now served by a continuation instead of being returned with an unresolvable `headroom_retrieve`. - Kill switch / disable path: `--no-ccr` (`ccr_handle_responses=False`) disables CCR response handling entirely, as before. - Unsafe override required: no - Qualification impact: none on compression; one extra upstream continuation per mixed turn, the same cost a lone retrieve already pays. If the continuation request fails, the turn costs what it does today plus that failed request. - Rollback path: revert this commit. ## Review Readiness - [x] I have performed a self-review - [x] This PR is ready for human review ## Checklist - [x] My code follows the project's style guidelines - [x] I have performed a self-review of my code - [x] I have commented my code, particularly in hard-to-understand areas - [x] I have made corresponding changes to the documentation - [x] My changes generate no new warnings - [x] I have added tests that prove my fix is effective or that my feature works - [x] New and existing unit tests pass locally with my changes - [x] I did **not** edit `CHANGELOG.md` — it is generated by release-please from my Conventional Commit PR title (a CI guard enforces this) ## Screenshots (if applicable) N/A ## Additional Notes - The model does not see the client calls it dropped: the continuation carries the trimmed assistant turn (any text and thinking, plus the retrieve) and its tool_result. If its text announced a client call, the model re-issues it in the continuation, or decides not to. - If the model calls a mixed turn again in the last allowed retrieval round, the handler returns that turn as before (`RESIDUAL_CCR_SKIPPED_MIXED`). - Responses and Gemini still leak a mixed `headroom_retrieve` to the client. Fixing them needs the reasoning item or thought-signature pairing handled, which is out of scope here. - Headroom Desktop vendors this logic against 0.39.0 until a release carries it. 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: JerrettDavis <mxjerrett@gmail.com>
148 lines
5.6 KiB
Python
148 lines
5.6 KiB
Python
from __future__ import annotations
|
|
|
|
from headroom.ccr.tool_calls import (
|
|
CCRToolCall,
|
|
drop_tool_calls,
|
|
extract_tool_calls,
|
|
has_ccr_tool_calls,
|
|
parse_ccr_tool_calls,
|
|
tool_call_id_for_provider,
|
|
)
|
|
from headroom.ccr.tool_injection import CCR_TOOL_NAME
|
|
|
|
HASH = "abc123def456abc123def456"
|
|
|
|
|
|
def test_extract_tool_calls_handles_provider_shapes() -> None:
|
|
anthropic = {"content": [{"type": "tool_use", "id": "t1", "name": CCR_TOOL_NAME}]}
|
|
openai = {
|
|
"choices": [
|
|
{
|
|
"message": {
|
|
"tool_calls": [
|
|
{"id": "c1", "function": {"name": CCR_TOOL_NAME, "arguments": "{}"}}
|
|
]
|
|
}
|
|
}
|
|
]
|
|
}
|
|
google = {
|
|
"candidates": [
|
|
{"content": {"parts": [{"functionCall": {"name": CCR_TOOL_NAME, "args": {}}}]}}
|
|
]
|
|
}
|
|
responses = {"output": [{"type": "function_call", "name": CCR_TOOL_NAME}]}
|
|
|
|
assert len(extract_tool_calls(anthropic, "anthropic")) == 1
|
|
assert len(extract_tool_calls(openai, "openai")) == 1
|
|
assert len(extract_tool_calls(google, "google")) == 1
|
|
assert len(extract_tool_calls(responses, "openai_responses")) == 1
|
|
|
|
|
|
def test_extract_tool_calls_rejects_invalid_shapes() -> None:
|
|
assert extract_tool_calls({"content": "not-a-list"}, "anthropic") == []
|
|
assert extract_tool_calls({"choices": []}, "openai") == []
|
|
assert extract_tool_calls({"choices": ["bad"]}, "openai") == []
|
|
assert extract_tool_calls({"candidates": [{"content": {"parts": "bad"}}]}, "google") == []
|
|
assert extract_tool_calls({"output": "bad"}, "openai_responses") == []
|
|
assert extract_tool_calls({}, "unknown") == []
|
|
|
|
|
|
def test_has_ccr_tool_calls_uses_provider_native_names() -> None:
|
|
assert has_ccr_tool_calls(
|
|
{"content": [{"type": "tool_use", "name": CCR_TOOL_NAME, "input": {"hash": HASH}}]},
|
|
"anthropic",
|
|
)
|
|
assert not has_ccr_tool_calls(
|
|
{"content": [{"type": "tool_use", "name": "read_file", "input": {"hash": HASH}}]},
|
|
"anthropic",
|
|
)
|
|
|
|
|
|
def test_ccr_detection_survives_null_function_tool_call() -> None:
|
|
# A partial/streamed OpenAI tool call with an explicit {"function": null}
|
|
# must not crash detection: dict.get("function", {}) returns None for a
|
|
# present-but-null key, and .get on None raises AttributeError.
|
|
response = {
|
|
"choices": [
|
|
{
|
|
"message": {
|
|
"tool_calls": [
|
|
{"id": "call_1", "type": "function", "function": None},
|
|
{
|
|
"id": "call_2",
|
|
"type": "function",
|
|
"function": {
|
|
"name": CCR_TOOL_NAME,
|
|
"arguments": '{"hash": "' + HASH + '"}',
|
|
},
|
|
},
|
|
]
|
|
}
|
|
}
|
|
]
|
|
}
|
|
|
|
assert has_ccr_tool_calls(response, "openai")
|
|
ccr_calls, other_calls = parse_ccr_tool_calls(response, "openai")
|
|
assert ccr_calls == [CCRToolCall(tool_call_id="call_2", hash_key=HASH)]
|
|
assert other_calls == [{"id": "call_1", "type": "function", "function": None}]
|
|
|
|
|
|
def test_parse_ccr_tool_calls_splits_retrievals_from_other_tools() -> None:
|
|
response = {
|
|
"content": [
|
|
{"type": "tool_use", "id": "tool_1", "name": CCR_TOOL_NAME, "input": {"hash": HASH}},
|
|
{"type": "tool_use", "id": "tool_2", "name": "read_file", "input": {"path": "a.py"}},
|
|
]
|
|
}
|
|
|
|
ccr_calls, other_calls = parse_ccr_tool_calls(response, "anthropic")
|
|
|
|
assert ccr_calls == [CCRToolCall(tool_call_id="tool_1", hash_key=HASH)]
|
|
assert other_calls == [
|
|
{"type": "tool_use", "id": "tool_2", "name": "read_file", "input": {"path": "a.py"}}
|
|
]
|
|
|
|
|
|
def test_tool_call_id_for_provider_models_matching_result_ids() -> None:
|
|
assert (
|
|
tool_call_id_for_provider({"functionCall": {"name": CCR_TOOL_NAME}}, "google")
|
|
== CCR_TOOL_NAME
|
|
)
|
|
assert (
|
|
tool_call_id_for_provider({"id": "item_1", "call_id": "call_1"}, "openai_responses")
|
|
== "call_1"
|
|
)
|
|
assert tool_call_id_for_provider({"id": "tool_1"}, "anthropic") == "tool_1"
|
|
|
|
|
|
def test_drop_tool_calls_removes_only_the_given_calls() -> None:
|
|
keep = {"type": "tool_use", "id": "r", "name": CCR_TOOL_NAME}
|
|
drop = {"type": "tool_use", "id": "b", "name": "Read"}
|
|
anthropic = {"content": [{"type": "text", "text": "hi"}, keep, drop], "stop_reason": "x"}
|
|
out = drop_tool_calls(anthropic, "anthropic", [drop])
|
|
assert out["content"] == [{"type": "text", "text": "hi"}, keep]
|
|
assert out["stop_reason"] == "x"
|
|
assert len(anthropic["content"]) == 3
|
|
|
|
call = {"id": "c", "function": {"name": "Read", "arguments": "{}"}}
|
|
openai = {"choices": [{"message": {"tool_calls": [call]}}, {"index": 1}]}
|
|
out = drop_tool_calls(openai, "openai", [call])
|
|
assert out["choices"] == [{"message": {"tool_calls": []}}, {"index": 1}]
|
|
assert openai["choices"][0]["message"]["tool_calls"] == [call]
|
|
|
|
|
|
def test_drop_tool_calls_returns_unrecognized_shapes_unchanged() -> None:
|
|
"""Callers detect "cannot drop here" by identity."""
|
|
for response, provider in [
|
|
({"content": "text"}, "anthropic"),
|
|
({"choices": []}, "openai"),
|
|
({"choices": [None]}, "openai"),
|
|
({"choices": [{"message": None}]}, "openai"),
|
|
({"choices": [{"message": {"content": "no tools"}}]}, "openai"),
|
|
({"output": []}, "openai_responses"),
|
|
({"candidates": []}, "google"),
|
|
]:
|
|
assert drop_tool_calls(response, provider, []) is response
|