feat: multi-level agent management — background agents, lifecycle tools, 3-level hierarchy

agent_manager.py (new):
- AgentRecord dataclass: agent_id, level (1/2/3), role, task, status, started,
  parent_id (lineage), finished, result, notify, _task_ref
- register() / finish() / cancel_agent() / list_agents() / get() / set_task_ref()
- Calls notification.notify() on completion when notify=True (same channel as
  reminders and cron completions)
- 24-hour pruning of completed records on each new registration

spawn_agent (tools/agents.py):
- background=True: fires asyncio.create_task(), registers in agent_manager, returns
  agent_id string immediately — sync path unchanged (no regression)
- notify=True: push/Talk notification when the background task completes
- Level enforcement: _agent_level param tracks hierarchy depth; when spawning from
  Level 2, child automatically gets spawn_agent + aider_run denied so Level 3 agents
  cannot delegate further

New lifecycle tools (tools/agents.py + __init__.py):
- agent_status(agent_id) — status, role, level, elapsed, task, result preview; user-level
- agent_list(status, limit) — all agents for current user, newest first; user-level
- agent_cancel(agent_id) — kills background task; admin-only, confirm-required

tests/test_agent_manager.py (new, 41 tests):
- agent_manager CRUD, pruning, notification hook
- spawn_agent background: returns immediately, completes async, timeout, failure
- Level enforcement: L1→L2 permits spawn, L2→L3 auto-denies; explicit tool_list path
- agent_status / agent_list / agent_cancel output formatting
- aider_run background: returns agent_id, completes async, sync path unchanged
- All tests run without browser or Cortex service (~2.5s total)
  Run: cd cortex && .venv/bin/python -m pytest tests/test_agent_manager.py -v

Docs: ARCH__FUTURE.md §13 (full design), ROADMAP.md, TODO__Agents.md, MASTER.md,
HELP.md (orchestrator description corrected, tool schema line updated to reflect
keyword routing), CLAUDE.md tool count 66→69.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

cortex/static/HELP.md

@@ -70,8 +70,8 @@ Click the **⚡** button in the input row to enable the Tools toggle. When lit (
 The orchestrator runs a multi-step tool loop:
 
 1. The **orchestrator model** reasons about the request and calls tools as needed
-2. It produces an enriched summary of what it found
-3. The **responder model** (set by the active Role) receives that context and writes the final user-facing reply
+2. Tool results are fed back into the conversation; the loop continues until the model has what it needs
+3. The model produces the final user-facing reply — when the orchestrator role uses Gemini, Claude writes the final response; when it uses a local model, that same model writes it
 4. Expandable tool-call cards appear above the response — click any card to see the arguments sent and the result returned
 
 The ⚡ toggle is **independent of the Role selector** — you can use any role (chat, coder, research, etc.) with or without tools. The orchestrator model is configured in **Account → Model Registry → Role Assignments → Orchestrator**.
@@ -82,7 +82,7 @@ Orchestrated sessions persist to history exactly like regular chat.
 
 ### Available Tools
 
-65 tools across 17 categories. Each tool schema is sent to the model on every orchestrated call — fewer active tools means fewer tokens per call.
+69 tools across 17 categories. Tool schemas are narrowed per-message using keyword routing — only categories relevant to your request are sent, keeping token overhead low. Per-role tool sets provide additional filtering.
 
 | Category | Tools |
 |---|---|
@@ -101,13 +101,14 @@ Orchestrated sessions persist to history exactly like regular chat.
 | **Aether Tasks** | `ae_task_list` |
 | **Aether Database** (admin) | `ae_db_query`, `ae_db_describe`, `ae_db_show_view` |
 | **Agent Notes** | `agent_notes_read`, `agent_notes_write`, `agent_notes_append`, `agent_notes_clear` |
-| **Agents** | `spawn_agent` |
+| **Agents** | `spawn_agent`, `aider_run` |
 | **Home Assistant** | `ha_get_state`, `ha_get_states`, `ha_call_service` |
 
 Files, Shell, System, Aether Database, Agents, and some Notification/Web tools are **admin-only** and not visible to regular users.
 `http_post` requires a URL prefix allowlist in `home/{user}/http_allowlist.json`.
 `nc_talk_history` requires `nc_username` and `nc_app_password` in `channels.json` under `nextcloud`.
 `ae_db_*` tools require Aether DB credentials configured in **Integrations** settings. All queries are SELECT-only — no writes possible.
+`aider_run` requires Aider installed (`pip install aider-chat`) and a model configured via `AIDER_MODEL` env var or the project's `.aider.conf.yml`. Supports any OpenAI-compatible backend — DeepSeek, OpenRouter, Ollama, etc.
 
 ### Per-Role Tool Sets