refactor(proxy): isolate output effort policy (#1961)

## Description

Extracts provider-neutral output effort decisions into a pure
`output_effort_policy` module. `output_shaper` still owns request
mutation and labels, while the rank comparisons, legacy thinking clamp,
and OpenAI text verbosity eligibility now live behind small
deterministic functions.

Closes #

## Type of Change

- [ ] 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
- [x] Code refactoring (no functional changes)

## Changes Made

- Added `headroom.proxy.output_effort_policy` for effort lowering,
legacy thinking budget clamping, and OpenAI text verbosity decisions.
- Updated `output_shaper` to delegate those pure decisions while
preserving existing labels and request mutation behavior.
- Added focused policy tests for effort rank transitions, thinking clamp
boundaries, and verbosity creation/lowering.
- Included the current LiteLLM callback signature compatibility shim
required for repo-wide mypy on main-based slices.

## Testing

- [x] Unit tests pass (`pytest`)
- [x] Linting passes (`ruff check .`)
- [x] Type checking passes (`mypy headroom`)
- [x] New tests added for new functionality
- [ ] Manual testing performed

### Test Output

```text
python -m pytest tests/test_output_effort_policy.py tests/test_output_shaper.py tests/test_litellm_callback.py -q
56 passed in 6.34s

python -m ruff check .
All checks passed!

python -m ruff format --check .
1095 files already formatted

python -m mypy headroom --ignore-missing-imports
Success: no issues found in 409 source files

gitleaks protect --staged --no-banner --redact
no leaks found
```

## Real Behavior Proof

- Environment: Windows, Python 3.13.13, clean worktree based on
`headroomlabs/main`.
- Exact command / steps: targeted pytest, ruff, format check, repo-wide
mypy, staged gitleaks scan.
- Observed result: output effort policy/shaper/callback tests pass;
static checks pass; no staged secrets detected.
- Not tested: live provider calls; this slice preserves existing request
mutation behavior and only moves pure policy decisions.

## 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
- [ ] 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
- [ ] I have updated the CHANGELOG.md if applicable

## Screenshots (if applicable)

N/A

## Additional Notes

Documentation and changelog updates are not applicable for this internal
architecture slice. PR-specific GHAS checks will be monitored after
opening.

Author: JerrettDavis

headroom/proxy/output_effort_policy.py

@@ -0,0 +1,55 @@
+"""Pure output-effort policy decisions.
+
+The output shaper mutates provider request bodies. This module owns the
+provider-neutral decisions behind those mutations so rank comparisons and
+legacy budget clamping stay testable without request dictionaries.
+"""
+
+from __future__ import annotations
+
+EFFORT_RANK = {"low": 0, "medium": 1, "high": 2, "xhigh": 3, "max": 4}
+TEXT_VERBOSITY_RANK = {"low": 0, "medium": 1, "high": 2}
+LEGACY_THINKING_FLOOR = 1024
+
+
+def lower_effort_value(current: object, target: str) -> str | None:
+    """Return ``target`` when an existing effort should be lowered."""
+    if not isinstance(current, str):
+        return None
+    if current not in EFFORT_RANK or target not in EFFORT_RANK:
+        return None
+    if EFFORT_RANK[current] <= EFFORT_RANK[target]:
+        return None
+    return target
+
+
+def clamp_legacy_thinking_budget(
+    *,
+    thinking_type: object,
+    budget_tokens: object,
+    floor: int = LEGACY_THINKING_FLOOR,
+) -> int | None:
+    """Return the clamped budget for legacy enabled thinking, else ``None``."""
+    if thinking_type != "enabled":
+        return None
+    if not isinstance(budget_tokens, int):
+        return None
+    if budget_tokens <= floor:
+        return None
+    return floor
+
+
+def can_create_openai_text_verbosity(model: object) -> bool:
+    """Whether it is safe to create a new OpenAI ``text.verbosity`` block."""
+    return str(model or "").lower().startswith("gpt-5")
+
+
+def lower_text_verbosity_value(current: object) -> str | None:
+    """Return ``low`` when an existing OpenAI text verbosity should be lowered."""
+    if not isinstance(current, str):
+        return None
+    if current not in TEXT_VERBOSITY_RANK:
+        return None
+    if TEXT_VERBOSITY_RANK[current] <= TEXT_VERBOSITY_RANK["low"]:
+        return None
+    return "low"

headroom/proxy/output_shaper.py

@@ -39,6 +39,16 @@
 from typing import Any
 
 from headroom.proxy import runtime_env
+from headroom.proxy.output_effort_policy import (
+    EFFORT_RANK as _EFFORT_RANK,
+)
+from headroom.proxy.output_effort_policy import (
+    LEGACY_THINKING_FLOOR,
+    can_create_openai_text_verbosity,
+    clamp_legacy_thinking_budget,
+    lower_effort_value,
+    lower_text_verbosity_value,
+)
 from headroom.proxy.output_steering import (
     apply_openai_responses_verbosity_steering,
     apply_verbosity_steering,
@@ -66,15 +76,6 @@
     "steering_text",
 ]
 
-# Documented Anthropic API minimum for thinking.budget_tokens on models
-# that still accept the legacy enabled/budget_tokens form.
-LEGACY_THINKING_FLOOR = 1024
-
-# Ordering for output_config.effort values. Unknown values are left alone.
-_EFFORT_RANK = {"low": 0, "medium": 1, "high": 2, "xhigh": 3, "max": 4}
-
-_TEXT_VERBOSITY_RANK = {"low": 0, "medium": 1, "high": 2}
-
 _OPENAI_RESPONSES_OUTPUT_ITEM_TYPES = frozenset(
     {
         "custom_tool_call_output",
@@ -265,22 +266,24 @@ def route_effort(
     output_config = body.get("output_config")
     if isinstance(output_config, dict):
         effort = output_config.get("effort")
-        if (
-            isinstance(effort, str)
-            and effort in _EFFORT_RANK
-            and _EFFORT_RANK[effort] > _EFFORT_RANK[settings.mechanical_effort]
-        ):
-            output_config["effort"] = settings.mechanical_effort
-            labels.append(f"output_shaper:effort:{effort}->{settings.mechanical_effort}")
+        lowered = lower_effort_value(effort, settings.mechanical_effort)
+        if lowered is not None:
+            output_config["effort"] = lowered
+            labels.append(f"output_shaper:effort:{effort}->{lowered}")
 
     # Legacy lever: clamp thinking.budget_tokens on models still using the
     # enabled/budget_tokens form. The type field itself is never touched.
     thinking = body.get("thinking")
-    if isinstance(thinking, dict) and thinking.get("type") == "enabled":
+    if isinstance(thinking, dict):
         budget = thinking.get("budget_tokens")
-        if isinstance(budget, int) and budget > LEGACY_THINKING_FLOOR:
-            thinking["budget_tokens"] = LEGACY_THINKING_FLOOR
-            labels.append(f"output_shaper:thinking_budget:{budget}->{LEGACY_THINKING_FLOOR}")
+        clamped = clamp_legacy_thinking_budget(
+            thinking_type=thinking.get("type"),
+            budget_tokens=budget,
+            floor=LEGACY_THINKING_FLOOR,
+        )
+        if clamped is not None:
+            thinking["budget_tokens"] = clamped
+            labels.append(f"output_shaper:thinking_budget:{budget}->{clamped}")
 
     return labels
 
@@ -363,22 +366,17 @@ def route_openai_reasoning_effort(
         return []
     effort = reasoning.get("effort")
     target = settings.mechanical_effort
-    if (
-        isinstance(effort, str)
-        and effort in _EFFORT_RANK
-        and target in _EFFORT_RANK
-        and _EFFORT_RANK[effort] > _EFFORT_RANK[target]
-    ):
-        reasoning["effort"] = target
-        return [f"output_shaper:reasoning_effort:{effort}->{target}"]
+    lowered = lower_effort_value(effort, target)
+    if lowered is not None:
+        reasoning["effort"] = lowered
+        return [f"output_shaper:reasoning_effort:{effort}->{lowered}"]
     return []
 
 
 def route_openai_text_verbosity(body: dict[str, Any]) -> list[str]:
     """Set or lower OpenAI ``text.verbosity`` conservatively."""
-    model = str(body.get("model") or "").lower()
     text_config = body.get("text")
-    can_create = model.startswith("gpt-5")
+    can_create = can_create_openai_text_verbosity(body.get("model"))
     if text_config is None:
         if not can_create:
             return []
@@ -393,13 +391,10 @@ def route_openai_text_verbosity(body: dict[str, Any]) -> list[str]:
             return []
         text_config["verbosity"] = "low"
         return ["output_shaper:text_verbosity:unset->low"]
-    if (
-        isinstance(verbosity, str)
-        and verbosity in _TEXT_VERBOSITY_RANK
-        and _TEXT_VERBOSITY_RANK[verbosity] > _TEXT_VERBOSITY_RANK["low"]
-    ):
-        text_config["verbosity"] = "low"
-        return [f"output_shaper:text_verbosity:{verbosity}->low"]
+    lowered = lower_text_verbosity_value(verbosity)
+    if lowered is not None:
+        text_config["verbosity"] = lowered
+        return [f"output_shaper:text_verbosity:{verbosity}->{lowered}"]
     return []
 
 

tests/test_output_effort_policy.py

@@ -0,0 +1,58 @@
+"""Tests for pure output effort policy decisions."""
+
+from __future__ import annotations
+
+from headroom.proxy.output_effort_policy import (
+    LEGACY_THINKING_FLOOR,
+    can_create_openai_text_verbosity,
+    clamp_legacy_thinking_budget,
+    lower_effort_value,
+    lower_text_verbosity_value,
+)
+
+
+def test_lower_effort_value_lowers_known_higher_effort_to_target() -> None:
+    assert lower_effort_value("xhigh", "low") == "low"
+    assert lower_effort_value("max", "medium") == "medium"
+
+
+def test_lower_effort_value_keeps_lower_equal_unknown_or_non_string_values() -> None:
+    assert lower_effort_value("low", "medium") is None
+    assert lower_effort_value("medium", "medium") is None
+    assert lower_effort_value("turbo", "low") is None
+    assert lower_effort_value("high", "turbo") is None
+    assert lower_effort_value(None, "low") is None
+
+
+def test_clamp_legacy_thinking_budget_only_clamps_enabled_over_floor() -> None:
+    assert (
+        clamp_legacy_thinking_budget(
+            thinking_type="enabled",
+            budget_tokens=32_000,
+        )
+        == LEGACY_THINKING_FLOOR
+    )
+    assert (
+        clamp_legacy_thinking_budget(
+            thinking_type="enabled",
+            budget_tokens=LEGACY_THINKING_FLOOR,
+        )
+        is None
+    )
+    assert clamp_legacy_thinking_budget(thinking_type="adaptive", budget_tokens=32_000) is None
+    assert clamp_legacy_thinking_budget(thinking_type="enabled", budget_tokens="32000") is None
+
+
+def test_can_create_openai_text_verbosity_only_for_gpt5_family() -> None:
+    assert can_create_openai_text_verbosity("gpt-5")
+    assert can_create_openai_text_verbosity("GPT-5.1")
+    assert not can_create_openai_text_verbosity("gpt-4o")
+    assert not can_create_openai_text_verbosity(None)
+
+
+def test_lower_text_verbosity_value_lowers_existing_verbose_values() -> None:
+    assert lower_text_verbosity_value("medium") == "low"
+    assert lower_text_verbosity_value("high") == "low"
+    assert lower_text_verbosity_value("low") is None
+    assert lower_text_verbosity_value("chatty") is None
+    assert lower_text_verbosity_value(None) is None