docs(progress): update Phase F status — 38/41 tasks done

PROGRESS.md

@@ -1,7 +1,7 @@
 # LeoPlatform — Fortschritts-Tracking
 
 > **Letztes Update:** 2026-08-17
-> **Status:** Phase B — done, Phase F startet
+> **Status:** Phase F — fast done (3 Tasks offen)
 
 ---
 
@@ -15,13 +15,13 @@
 | C.5 — Import/Export | `done` | 2026-08-13 | 2026-08-13 | 8 | 8 |
 | D — Undo/Restore | `done` | 2026-08-13 | 2026-08-13 | 13 | 13 |
 | E — Search | `done` | 2026-08-14 | 2026-08-14 | 24 | 24 |
-| F — Agents | `in_progress` | 2026-08-17 | — | 1 | ~28 |
+| F — Agents | `in_progress` | 2026-08-17 | — | ~38 | ~41 |
 | G — Workflows | `not_started` | — | — | 0 | ~24 |
 | H — Knowledge | `not_started` | — | — | 0 | ~18 |
 | I — Integration & Workstream | `not_started` | — | — | 0 | ~25 |
 | J — Self-Improvement | `not_started` | — | — | 0 | ~12 |
 
-**Gesamt:** ~114 / ~223 Tasks done
+**Gesamt:** ~152 / ~223 Tasks done
 
 ---
 
@@ -231,6 +231,56 @@
 
 ---
 
+## Phase F — Agent MVP
+
+| Task | Status | Forgejo Issue | Verifiziert |
+|------|-------|---------------|------------|
+| F-LOOP | `done` | — | ✅ app/ai/agent_loop.py — ReActStep/ReActResult + run_react_loop() mit LLM→Tool→Observe Loop, max_steps/timeout Graceful-Stop, ErrorCategory Retry, Cost-Accumulation, agent.step Hook. 11/11 Tests grün |
+| F-CALL | `done` | — | ✅ Tool-Call-Parser in agent_loop.py (_extract_tool_calls) |
+| F-MAX | `done` | — | ✅ Max-Steps Limit + Graceful Stop in agent_loop.py |
+| F-ERR | `done` | — | ✅ ErrorCategory handling (TRANSIENT→retry, PERMANENT→stop, PARTIAL→continue) |
+| F-CTX | `done` | — | ✅ app/ai/context_builder.py (282 lines) — build_agent_context() + ReActSystemPromptBuilder |
+| F-STR | `done` | — | ✅ app/ai/agent_stream.py (155 lines) — stream_react_loop() with SSE events |
+| F-DEF | `done` | — | ✅ AgentDefinition fields (temperature, max_tokens, max_steps, trace_mode, skill_ids, trigger_config, ai_use_case_metadata) + migration 0122 |
+| F-SKILL | `done` | — | ✅ app/ai/skill_registry.py (82 lines) — SkillDefinition + SkillRegistry singleton |
+| F-TOOL | `done` | — | ✅ app/ai/agent_tools.py (117 lines) — get_agent_tools() with permission intersection |
+| F-PERM | `done` | — | ✅ app/ai/agent_permissions.py (230 lines) — AgentPermissionContext, resolve_agent_permissions(), optimistic locking |
+| F-PERM-VIS | `done` | — | ✅ filter_visible_agents() in agent_permissions.py |
+| F-PERM-USE | `done` | — | ✅ check_agent_execute_permission() in agent_permissions.py |
+| F-DRY | `done` | — | ✅ Dry-Run Mode in agent_loop.py |
+| F-AUDIT | `done` | — | ✅ Audit-Log für Tool-Calls in agent_loop.py |
+| F-AIUSE | `done` | — | ✅ app/ai/ai_use_case.py (156 lines) — AIUseCaseMetadata, validate_ai_use_case() |
+| F-TRANS | `done` | — | ✅ app/ai/transparency.py (60 lines) — mark_as_ai_generated(), is_ai_participant() |
+| F-DATA-POL | `done` | — | ✅ app/ai/data_policy.py (210 lines) — enforce_data_policy() with SENSITIVE_FIELDS + provider compliance |
+| F-OVERSIGHT | `done` | — | ✅ app/ai/oversight.py (108 lines) — DecisionRecord, create_decision_record() |
+| F-APPR | `done` | — | ✅ app/core/approval.py (160 lines) + app/routes/approvals.py (305 lines) + migration 0123 — ApprovalRequest CRUD API |
+| F-MEM | `done` | — | ✅ app/ai/agent_memory.py (236 lines) — store/retrieve/search agent memory with embeddings |
+| F-PROACTIVE | `done` | — | ✅ trigger_dispatcher.py — _dispatch_matching_agents() for context/UI events |
+| F-WORK | `done` | — | ✅ app/ai/agent_workstream.py (201 lines) — post_agent_message, post_agent_step, post_agent_result, post_approval_request |
+| F-UI-CTRL | `done` | — | ✅ Bestehendes AI UI Control Plugin (ai_ui_control) |
+| F-UI-LIST | `done` | — | ✅ AgentDashboard.tsx (831 lines), AgentsOverview.tsx |
+| F-UI-EDIT | `done` | — | ✅ AgentEditor.tsx (396 lines) |
+| F-UI-CHAT | `done` | — | ✅ AgentChat.tsx (147 lines) — SSE streaming, extended trace, dry run, cost display |
+| F-UI-LOG | `done` | — | ✅ AgentRunLog.tsx (82 lines) — timeline view, export JSON/CSV |
+| F-UI-MON | `done` | — | ✅ AgentMonitor.tsx (86 lines) — live stats, auto-refresh 5s |
+| F-EMAIL | `done` | — | ✅ prebuilt/email_triage_agent.py — E-Mail-Triage-Agent |
+| F-CONTACT | `done` | — | ✅ prebuilt/contact_enrichment_agent.py — Contact-Enrichment-Agent |
+| F-FOLLOW | `done` | — | ✅ prebuilt/follow_up_agent.py — Follow-up-Agent |
+| F-REPORT | `done` | — | ✅ prebuilt/report_agent.py — Report-Agent |
+| F-UI-TRIG | `not_started` | — | — |
+| F-TASK-MODEL | `done` | — | ✅ Extended Task model with polymorphic assignee/entity/creator, subtasks, dependencies, task_type, success_criteria, progress |
+| F-TASK-API | `done` | — | ✅ Extended task routes with polymorphic filters, subtasks, dependencies |
+| F-TASK-AGENT | `done` | — | ✅ ai_tools.py (191 lines) — create_task, assign_task, update_task_status, decompose_goal tools |
+| F-TASK-WORK | `done` | — | ✅ workstream.py — task_card and goal_card blocks |
+| F-TASK-UI | `done` | — | ✅ TaskBoard.tsx (147 lines), TaskDetail.tsx (287 lines), GoalView.tsx (152 lines) |
+| F-TASK-MIG | `done` | — | ✅ Migration 0124 — new columns, data migration |
+| F-TASK-GOAL | `done` | — | ✅ Progress aggregation, success criteria evaluation, parent status propagation |
+| F-TASK-TEST | `done` | — | ✅ test_unified_tasks.py (414 lines) |
+| F-TEST | `not_started` | — | — |
+| F-DOC | `not_started` | — | — |
+
+---
+
 ## Phasen E-J
 
 Detaillierte Task-Listen werden beim Start der jeweiligen Phase eingetragen. Siehe `PLATFORM_ROADMAP.md` für alle Tasks.