Claude Opus 5.5 (effort default (no --effort flag) · thinking on) on Model ladder: fix 10 real merged bugs with tests hidden

[
 {
  "duration_api_ms": 40097,
  "stop_reason": "end_turn",
  "session_id": "[REDACTED]",
  "total_cost_usd": 0.2742442,
  "usage": {
   "input_tokens": 24,
   "cache_creation_input_tokens": 17045,
   "cache_read_input_tokens": 269741,
   "output_tokens": 4192,
   "output_tokens_details": {
    "thinking_tokens": 351
   },
   "server_tool_use": {
    "web_search_requests": 0,
    "web_fetch_requests": 0
   },
   "service_tier": "standard",
   "cache_creation": {
    "ephemeral_1h_input_tokens": 17045,
    "ephemeral_5m_input_tokens": 0
   },
   "inference_geo": "global",
   "iterations": [
    {
     "input_tokens": 2,
     "output_tokens": 559,
     "cache_read_input_tokens": 28675,
     "cache_creation_input_tokens": 516,
     "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 516
     },
     "type": "message"
    }
   ],
   "speed": "standard",
   "fallback_credit": null
  },
  "modelUsage": {
   "claude-opus-5-5": {
    "inputTokens": 24,
    "outputTokens": 4192,
    "cacheReadInputTokens": 269741,
    "cacheCreationInputTokens": 17045,
    "webSearchRequests": 0,
    "costUSD": 0.2742442,
    "contextWindow": 1000000,
    "maxOutputTokens": 128000,
    "thinkingTokens": 351,
    "canonicalModel": "claude-opus-5-5",
    "provider": "firstParty",
    "costBasis": "list"
   }
  },
  "permission_denials": [],
  "terminal_reason": "completed",
  "fast_mode_state": "off",
  "fast_mode_disabled_reason": "sdk_opt_in_required",
  "subagent_stats": {
   "spawned": 0,
   "requested": {
    "background": 0,
    "foreground": 0,
    "unset": 0
   },
   "started_in_background": 0,
   "max_depth": 0,
   "spawned_by_subagents": 0,
   "completed": 0,
   "failed": 0,
   "killed": {
    "parent": 0,
    "user": 0,
    "system": 0
   },
   "refused": {
    "depth_limit": 0,
    "concurrency_limit": 0,
    "budget": 0
   },
   "by_type": {}
  },
  "safety_stops": 0,
  "is_error": false,
  "num_turns": 15,
  "subtype": "success",
  "api_error_status": null,
  "result": "I fixed the bug: a wait strategy that is callable is now always called, even if the object itself evaluates as false. The full test suite passes (185 passed, 1 skipped). I couldn't run a type check because mypy isn't installed in `.venv`.\n\n**Cause:** both the sync and async retry code decided whether to wait with `if not self.wait:`. That check was meant to catch the \"no wait\" values `None` and `0` (including the `0` you get from `sum([])` over an empty list of strategies). But it also caught real strategies that happen to be falsy, like your empty `list` subclass or an object whose `__bool__` returns False, so they were never called.\n\n**Fix:** the check is now `if not callable(self.wait) and not self.wait:` in both places:\n- `tenacity/__init__.py:428` (`BaseRetrying._run_wait`)\n- `tenacity/asyncio/__init__.py:159` (`AsyncRetrying._run_wait`)\n\n`None` and `0` still mean \"no wait\", since neither is callable. A truthy value that isn't callable still raises a `TypeError` when tenacity tries to call it, as it did before.\n\n**Tests added:**\n- `tests/test_tenacity.py`: `test_wait_falsy_callable_is_called` covers an empty `list` subclass and an object whose `__bool__` returns False. Each returns 1, and the test checks that `sleep(1)` is called between attempts.\n- `tests/test_asyncio.py`: the same test for `AsyncRetrying`, checking that the async sleep is awaited once with `1`.\n\nAll three new checks fail with the library change reverted and pass with it. The existing tests for `None` and `0` still pass.",
  "ttft_ms": 2572,
  "type": "result",
  "duration_ms": 44254,
  "uuid": "[REDACTED]",
  "ttft_stream_ms": 1284,
  "time_to_request_ms": 433,
  "first_content_frame_ms": 1284,
  "queued_turn_count": 0,
  "result_index": 0
 }
]