Feat/group task orchestration

AGENTS.md

@@ -105,7 +105,7 @@ Follow this order unless the user explicitly requests something different.
 | 7     | Security & Compliance    | Secrets Vault, Audit Ledger (EU AI Act), tenant quota stub                                          |
 | 8     | MCP Integration          | MCP Server (33 tools), MCP Client, agent discovery, managed conversations                           |
 | 8c    | RAG Foundation           | Config-driven vector store retrieval, pgvector, httpCall RAG                                        |
-| 10    | Group Conversations      | Multi-agent debate orchestration, 5 styles, group-of-groups                                         |
+| 10    | Group Conversations      | Multi-agent debate orchestration, 6 styles (incl. Task Force), group-of-groups                      |
 | —     | A2A Protocol             | Agent-to-Agent peer communication, Agent Cards, skill discovery                                     |
 | —     | Multi-Model Cascading    | Sequential model escalation with confidence routing                                                 |
 | —     | LLM Provider Expansion   | 7 → 12 providers (Mistral, Azure OpenAI, Bedrock, Oracle GenAI)                                     |

HANDOFF.md

@@ -555,18 +555,18 @@ All hooks are **non-fatal** — if schedule operations fail, the primary agent o
 
 ### Phase 10: Group Conversations — Multi-Agent Debate Orchestration ✅
 
-Phase-based orchestration engine enabling structured multi-agent discussions with 5 preset styles, nested group-of-groups support, and moderator synthesis.
+Phase-based orchestration engine enabling structured multi-agent discussions with 6 preset styles, nested group-of-groups support, and moderator synthesis.
 
 | Sub-Phase | Deliverables |
 |---|---|
 | **10.1** Data Models + Stores | `AgentGroupConfiguration`, `GroupConversation`, group/conversation stores (DB-agnostic) |
 | **10.2** Orchestration Service | `GroupConversationService` (~680 lines) — phases, context scoping, templates |
 | **10.3** REST + SSE + MCP | REST endpoints, SSE streaming, `McpGroupTools` (9 tools) |
-| **10.5** Discussion Styles | `DiscussionPhase`, `DiscussionStylePresets` — 5 styles + custom phases |
+| **10.5** Discussion Styles | `DiscussionPhase`, `DiscussionStylePresets` — 6 styles + custom phases |
 | **10.5b** MCP/REST Usability | `describe_discussion_styles`, `list_group_conversations`, `GET /styles` |
 | **10.6** Group-of-Groups | `MemberType.GROUP`, recursive `executeGroupMemberTurn()`, `memberTypes` param |
 
-**Discussion styles:** `ROUND_TABLE`, `PEER_REVIEW`, `DEVIL_ADVOCATE`, `DELPHI`, `DEBATE` (+ fully custom phases)
+**Discussion styles:** `ROUND_TABLE`, `PEER_REVIEW`, `DEVIL_ADVOCATE`, `DELPHI`, `DEBATE`, `TASK_FORCE` (+ fully custom phases)
 
 **Group-of-Groups:** Members can be other groups — sub-group synthesized answer becomes member's response. Depth tracking prevents infinite recursion (`eddi.groups.max-depth`, default: 3).
 

README.md

@@ -249,7 +249,7 @@ Most multi-agent frameworks (LangGraph, CrewAI, AutoGen) are Python/Node librari
 ### 🤖 Multi-Agent Orchestration
 
 - 🔀 **Intelligent Routing** — Direct conversations to different agents based on context, rules, and intent
-- 🗣️ **Group Conversations** — Multi-agent debates with 5 built-in discussion styles: Round Table, Peer Review, Devil's Advocate, Delphi, and Debate
+- 🗣️ **Group Conversations** — Multi-agent debates with 6 built-in discussion styles: Round Table, Peer Review, Devil's Advocate, Delphi, Debate, and Task Force
 - 💬 **Slack Integration** — Deploy agents to Slack channels and run multi-agent debates directly in threads
 - 🪆 **Nested Groups** — Compose groups of groups for tournament brackets, red-team vs blue-team, and panel reviews
 - 👥 **Managed Conversations** — Intent-based auto-routing with one conversation per user per intent

docs/architecture.md

@@ -868,7 +868,7 @@ A `GroupConversationService` orchestrates discussions through configurable phase
 
 **Key capabilities:**
 
-- **5 built-in discussion styles**: Round Table, Peer Review, Devil's Advocate, Delphi, and Debate — each with distinct phase flows and turn-taking rules
+- **6 built-in discussion styles**: Round Table, Peer Review, Devil's Advocate, Delphi, Debate, and Task Force — each with distinct phase flows and turn-taking rules. Task Force uses a 4-phase pipeline (PLAN→EXECUTE→VERIFY→SYNTHESIS) for structured task decomposition and parallel execution
 - **Custom phases**: Define your own phase sequences with configurable context scopes (independent, full transcript, anonymous, own-feedback-only)
 - **Group-of-groups**: Members can themselves be groups, enabling hierarchical multi-agent composition with configurable depth limits
 - **Fault tolerance**: Per-agent timeouts, configurable failure policies (skip, retry, abort), and graceful degradation when members are unavailable