fix(types): declare AIAgent instance attributes as class-level annotations

init_agent() in agent/agent_init.py sets ~176 instance attributes on the
AIAgent instance, but ty cannot track cross-module attribute assignment
through a function that receives the instance as a plain `agent` parameter.
This caused 186 unresolved-attribute errors in run_agent.py alone.

Declaring the attributes as class-level annotations (PEP 526) tells ty
they exist, clearing 182 of 190 errors (190→8). The remaining 8 are:
- 2 duck-typed object.function accesses (hasattr-guarded, safe)
- 2 missing attrs (_current_tool, _api_call_count — added)
- 2 None-narrowing on iteration_budget (.used, .max_total)
- 2 None-narrowing on client (.close)

Overall ty diagnostics: 4,648 → 4,422 (−226)
Tests: 496 passed, 0 failed

notes.txt

@@ -174,3 +174,90 @@ the pattern is always identical: `param: Type = None` → `param: Type | None =
 two runtime-breaking gotchas found during the sweep:
 1. lowercase `callable` (builtin function) → must be `Callable` (typing)
 2. string forward-refs can't use `| None` → must use `Optional["Ref"]`
+
+---
+
+## run_agent.py — unresolved-attribute analysis (190 errors)
+
+### category 1: "Self@<method>" — init_agent() cross-module init (186/190)
+
+these are ALL the same root cause: `AIAgent.__init__` in `run_agent.py` is a
+thin forwarder that calls `init_agent(self, ...)` from `agent/agent_init.py`.
+that function sets `self.model`, `self.provider`, `self.session_id`, etc. on
+the instance — but ty can't see across the module boundary that these
+attributes are being set. so every method that accesses `self.model`,
+`self.provider`, etc. gets flagged.
+
+**these are NOT bugs.** the attributes are correctly set at runtime by
+`init_agent()`. this is a ty limitation — it can't track attribute assignments
+made in a function defined in a different module that receives `self` as a
+regular parameter (not as a method on the class).
+
+example:
+```python
+# run_agent.py
+class AIAgent:
+    def __init__(self, model: str = "", ...):
+        from agent.agent_init import init_agent
+        init_agent(self, model=model, ...)  # sets self.model, self.provider, etc.
+
+    def _resolved_api_call_timeout(self):
+        return self.provider  # ty: "Self@_resolved_api_call_timeout has no attribute provider"
+```
+
+**fix approach:** this is the single biggest cluster of ty errors in the
+codebase (186 in run_agent.py alone, plus similar in other files). the cleanest
+fix would be to declare the attributes as class-level annotations in the
+`AIAgent` class body so ty knows they exist:
+
+```python
+class AIAgent:
+    # Instance attributes — set by init_agent() in agent/agent_init.py
+    model: str
+    provider: str | None
+    session_id: str | None
+    # ... etc
+```
+
+this is a one-time declaration that would clear ~186 errors instantly. it's
+also good documentation — currently there's no single place that lists all
+instance attributes; they're scattered across the 60-param init_agent() body.
+
+### category 2: object.function — duck-typed tool_calls access (2 errors)
+
+line 1968: `tc.function.name` and `tc.function.arguments` on objects typed as
+`object`. this is because `msg.tool_calls` is checked via `hasattr` +
+`isinstance(list)` but the list elements are `object` to ty.
+
+```python
+if hasattr(msg, "tool_calls") and isinstance(msg.tool_calls, list) and msg.tool_calls:
+    tool_calls_data = [
+        {"name": tc.function.name, "arguments": tc.function.arguments}
+        for tc in msg.tool_calls
+    ]
+```
+
+**not a bug** — this is duck-typed access to OpenAI SDK objects. the `hasattr`
+guard makes it safe at runtime. ty just can't infer the element type of a
+list accessed via `hasattr`. could be fixed with a `type: ignore` or by
+narrowing `msg` to a proper type.
+
+### category 3: ~AlwaysFalsy.on_session_end — duck-typed context_compressor (2 errors)
+
+lines 3416, 3441: `self.context_compressor.on_session_end(...)` after a
+`hasattr(self, "context_compressor") and self.context_compressor` guard.
+
+```python
+if hasattr(self, "context_compressor") and self.context_compressor:
+    self.context_compressor.on_session_end(self.session_id or "", messages or [])
+```
+
+ty narrows `self.context_compressor` to `~AlwaysFalsy` (the truthy branch of
+the `and`) but doesn't know it has `on_session_end`. **not a bug** — the
+`hasattr` guard makes this safe. again a type-system limitation with
+duck-typed access.
+
+### verdict for run_agent.py: 0 real bugs, 190 type-system limitations
+
+all 190 are ty being unable to track either cross-module attribute init
+(186) or duck-typed hasattr-guarded access (4). no logic bugs found.

run_agent.py

@@ -400,6 +400,200 @@ class AIAgent:
     for AI models that support function calling.
     """
 
+    # -----------------------------------------------------------------------
+    # Instance attributes — set by init_agent() in agent/agent_init.py.
+    # Ideally, we could refactor this class into smaller parts so that
+    # we don't need to split its __init__ into a separate file.
+    # Declared here so type checkers know they exist (init_agent receives
+    # the instance as a plain `agent` parameter, not as a method receiver,
+    # so cross-module attribute assignment is invisible to static analysis).
+    # -----------------------------------------------------------------------
+    # scalars
+    _anthropic_api_key: str
+    _anthropic_base_url: str | None
+    _budget_exhausted_injected: bool
+    _budget_grace_call: bool
+    _cache_ttl: str
+    _cached_system_prompt: str | None
+    _chat_id: str | None
+    _chat_name: str | None
+    _chat_type: str | None
+    _client_kwargs: dict
+    _codex_reasoning_replay_enabled: bool
+    _compression_feasibility_checked: bool
+    _credits_latch: dict
+    _current_streamed_assistant_text: str
+    _delegate_depth: int
+    _end_session_on_close: bool
+    _executing_tools: bool
+    _fallback_activated: bool
+    _fallback_chain: list
+    _fallback_index: int
+    _force_ascii_payload: bool
+    _interrupt_requested: bool
+    _interrupt_thread_signal_pending: bool
+    _is_anthropic_oauth: bool
+    _is_user_initiated_turn: bool
+    _iters_since_skill: int
+    _last_flushed_db_idx: int
+    _memory_enabled: bool
+    _memory_nudge_interval: int
+    _memory_write_context: str
+    _memory_write_origin: str
+    _parent_session_id: str | None
+    _persist_disabled: bool
+    _primary_runtime: dict
+    _session_db_created: bool
+    _session_init_model_config: dict
+    _session_json_enabled: bool
+    _skill_nudge_interval: int
+    _skip_mcp_refresh: bool
+    _stream_needs_break: bool
+    _stream_writer_dropped: int
+    _stream_writer_token: int
+    _thread_id: str | None
+    _tool_snapshot_generation: int
+    _turns_since_memory: int
+    _user_name: str | None
+    _user_profile_enabled: bool
+    _user_turn_count: int
+    acp_command: str | None
+    api_key: str | None
+    api_mode: str | None
+    ephemeral_system_prompt: str | None
+    lmstudio_load_mode: str
+    load_soul_identity: bool
+    log_prefix: str
+    log_prefix_chars: int
+    max_iterations: int
+    max_tokens: int | None
+    memory_notifications: str
+    model: str
+    pass_session_id: bool
+    platform: str | None
+    provider: str | None
+    provider_data_collection: str | None
+    provider_require_parameters: bool
+    provider_sort: str | None
+    quiet_mode: bool
+    save_trajectories: bool
+    service_tier: str | None
+    session_api_calls: int
+    session_cache_read_tokens: int
+    session_cache_write_tokens: int
+    session_completion_tokens: int
+    session_cost_source: str
+    session_cost_status: str
+    session_estimated_cost_usd: float
+    session_id: str | None
+    session_input_tokens: int
+    session_output_tokens: int
+    session_prompt_tokens: int
+    session_reasoning_tokens: int
+    session_total_tokens: int
+    show_commentary: bool
+    skip_context_files: bool
+    suppress_status_output: bool
+    tool_delay: float
+    tool_progress_mode: str
+    verbose_logging: bool
+
+    # callbacks
+    clarify_callback: Callable | None
+    event_callback: Optional[Callable[[str, dict], None]]
+    interim_assistant_callback: Callable | None
+    notice_callback: Callable | None
+    notice_clear_callback: Callable | None
+    reaction_callback: Optional[Callable[[str], None]]
+    read_terminal_callback: Callable | None
+    reasoning_callback: Callable | None
+    status_callback: Callable | None
+    step_callback: Callable | None
+    stream_delta_callback: Callable | None
+    thinking_callback: Callable | None
+    tool_complete_callback: Callable | None
+    tool_gen_callback: Callable | None
+    tool_progress_callback: Callable | None
+    tool_start_callback: Callable | None
+
+    # collections
+    acp_args: list[str] | None
+    disabled_toolsets: List[str] | None
+    enabled_toolsets: List[str] | None
+    prefill_messages: List[Dict[str, Any]] | None
+    providers_allowed: List[str] | None
+    providers_ignored: List[str] | None
+    providers_order: List[str] | None
+    reasoning_config: Dict[str, Any] | None
+    request_overrides: Dict[str, Any] | None
+
+    # internal / runtime state
+    _active_children: list
+    _active_children_lock: Any
+    _anthropic_client: Any | None
+    _anthropic_image_fallback_cache: Any
+    _api_max_retries: Any
+    _aux_compression_context_length_config: Any | None
+    _base_url_hostname: Any
+    _bedrock_guardrail_config: Any | None
+    _bedrock_region: Any
+    _checkpoint_mgr: Any
+    _client_lock: Any
+    _compression_threshold_autoraised: Any | None
+    _compression_warning: Any | None
+    _config_context_length: Any
+    _credential_pool: Any
+    _credits_session_start_micros: Any | None
+    _credits_state: Any | None
+    _custom_providers: Any
+    _environment_probe: Any
+    _execution_thread_id: Any
+    _fallback_model: Any
+    _gateway_session_key: Any
+    _intent_ack_continuation: Any
+    _interrupt_message: Any
+    _kanban_worker_guidance: Any
+    _memory_manager: Any | None
+    _memory_store: Any | None
+    _ollama_num_ctx: Any
+    _parallel_tool_call_guidance: Any
+    _pending_cli_user_message: Any | None
+    _pending_steer_lock: Any
+    _persist_user_message_idx: Any | None
+    _persist_user_message_override: Any | None
+    _persist_user_message_timestamp: Any | None
+    _platform_hint_overrides: Any
+    _print_fn: Any | None
+    _session_db: Any
+    _session_persist_lock: Any
+    _stream_callback: Any | None
+    _stream_context_scrubber: Any
+    _stream_think_scrubber: Any
+    _stream_writer_lock: Any
+    _stream_writer_tls: Any
+    _subdirectory_hints: Any
+    _task_completion_guidance: Any
+    _todo_store: Any
+    _tool_guardrails: Any
+    _tool_use_enforcement: Any
+    _tool_worker_threads_lock: Any
+    _user_id: Any
+    _user_id_alt: Any
+    background_review_callback: Any
+    client: Any | None
+    codex_app_server_auto_compaction: Any
+    compression_enabled: Any
+    compression_in_place: Any
+    context_compressor: Any
+    iteration_budget: Optional[IterationBudget]
+    logs_dir: Any
+    openrouter_min_coding_score: Optional[float]
+    session_start: Any
+    tools: Any
+    valid_tool_names: Any
+    _current_tool: Any
+    _api_call_count: int
+
     _TOOL_CALL_ARGUMENTS_CORRUPTION_MARKER = (
         "[hermes-agent: tool call arguments were corrupted in this session and "
         "have been dropped to keep the conversation alive. See issue #15236.]"