mindforge-cc 2.3.5 → 3.0.0

package/docs/commands-reference.md

@@ -1,4 +1,4 @@
-# MindForge v2.1.1 — Commands Reference
+# MindForge v2.4.0 — Commands Reference
 
 MindForge commands are organized into functional pillars to support the entire software development lifecycle (SDLC).
 
@@ -19,7 +19,7 @@ MindForge commands are organized into functional pillars to support the entire s
 
 | Command | Description |
 | :--- | :--- |
-| `/mindforge:plan-phase [N]` | Initiates strategic planning for a milestone: discuss → research → plan. |
+| `/mindforge:plan-phase [N]` | Initiates strategic planning for a milestone: discuss → research → plan. Supports `--ads` for adversarial synthesis. |
 | `/mindforge:execute-phase [N]` | Starts autonomous execution of task plans in parallel waves. |
 | `/mindforge:verify-phase [N]` | Runs Human Acceptance Testing (UAT) and automated validation gates. |
 | `/mindforge:ship [N]` | Finalizes delivery, generates release output, and creates a Pull Request. |
@@ -35,6 +35,7 @@ MindForge commands are organized into functional pillars to support the entire s
 | `/mindforge:personas --list` | Displays all 32+ specialized engineering personas. |
 | `/mindforge:personas --set ID` | Switches the current agent to a specific persona (e.g., `architect`, `executor`). |
 | `/mindforge:tokens` | Analyzes and profiles token usage for the current session. |
+| `shard-helper --verify` | (CLI Internal) Verifies the integrity and SHA-256 checksums of memory shards. |
 
 ---
 

package/docs/usp-features.md

@@ -1,4 +1,4 @@
-# MindForge v2.0.0 — Unique Selling Points, Features, and Best Practices (v2.0.0)
+# MindForge v3.0.0 — Unique Selling Points, Features, and Best Practices (v3.0.0)
 
 This document summarizes what makes MindForge v2.0.0 distinct, what features
 are included in the official release, and how to use them effectively.
@@ -40,15 +40,24 @@ are included in the official release, and how to use them effectively.
 10. **Multi-Model Intelligence Layer (v2)**
     - Dynamic routing between Anthropic, OpenAI, and Gemini based on task persona and security tier. Adversarial code reviews ensure maximum coverage.
 
-11. **Persistent Knowledge Graph (v2)**
-    - Captures and ranks engineering context (decisions, bug patterns, preferences) across project sessions using TF-IDF and confidence reinforcement.
+11. **Persistent Knowledge Graph (RAG 2.0)**
+    - Captures and ranks engineering context (decisions, bug patterns, preferences) with **Automatic Semantic Shadowing** — background pattern retrieval without manual prompts.
 
-12. **Real-time Observability Dashboard (v2)**
-    - High-fidelity web interface for live audit streams, metrics visualization, and browser-based governance with zero performance overhead.
+12. **Real-time Observability Dashboard (v3)**
+    - High-fidelity web interface for live audit streams, metrics visualization, and **Temporal History Scrubbing** with zero performance overhead.
 
-13. **Self-Building Skills Platform (v2)**
+13. **Self-Building Skills Platform (v3)**
     - Automatically capture and score engineering patterns from docs, npm, or session history into high-quality `SKILL.md` files with non-bypassable quality gates.
 
+14. **Context Sharding (v3)**
+    - 40% reduction in token waste via **Semantic Relevance Density (SRD)** management of Hot/Warm/Cold context tiers.
+
+15. **Adversarial Decision Synthesis (v3)**
+    - Zero architectural drift through Red-Team/Blue-Team debate loops and automated synthesis scoring.
+
+16. **Temporal Vision (v3)**
+    - Full-fidelity history navigation, hindsight injection, and automated state repair across the execution wave.
+
 ---
 
 ## Feature Set (v2.0.0)
@@ -206,25 +215,34 @@ preserving scope (local vs global).
 
 ---
 
-### 16. Real-time Dashboard (v2)
-**What it does:** Web-based control plane for observing agent waves, costs, and quality metrics in real-time.
+### 18. Adversarial Decision Synthesis (v3)
+**What it does:** Ensures architectural integrity by debating every major decision through competing model personas.
+
+**How to use:**
+```bash
+/mindforge:cross-review --ads # Triggers heavy synthesis mode
+```
+
+---
+
+### 19. Temporal Vision & Hindsight (v3)
+**What it does:** Provides the "Temporal Slider" for the Dashboard and the `hindsight-injector` for automated state repair.
 
 **How to use:**
 ```bash
-/mindforge:dashboard --start --open
+/mindforge:temporal status
+/mindforge:temporal inject [AUDIT_ID]
 ```
-Access at `http://localhost:7339` (Localhost-only for security).
 
 ---
 
-### 17. Self‑Building Skills Platform (v2)
-**What it does:** Automatically captures engineering knowledge into high-quality `SKILL.md` files. Features a 7-dimension quality scorer and npm-based marketplace integration.
+### 20. Context Sharding Manager (v3)
+**What it does:** Dramatically improves session longevity by sharding context based on sematic relevance.
 
 **How to use:**
 ```bash
-/mindforge:learn https://docs.example.com "my-skill"
-/mindforge:marketplace search "auth"
-/mindforge:marketplace install mindforge-skill-example
+/mindforge:memory shard
+/mindforge:memory status --srd
 ```
 
 ---

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "mindforge-cc",
-  "version": "2.3.5",
+  "version": "3.0.0",
   "description": "MindForge - Enterprise Agentic Framework for Claude Code and Antigravity",
   "bin": {
     "mindforge-cc": "bin/install.js"