@namch/agent-assistant 1.0.4 → 1.1.1

package/CHANGELOG.md

@@ -5,6 +5,52 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+## [1.1.1] - 2026-02-05
+
+### Fixed
+
+- **Path Placeholder Resolution**: Fixed critical issue where `{TOOL}`, `{TOOL}`, `~/.{TOOL}/` placeholders were not properly replaced during installation
+  - Added missing replacement patterns to all tools in `cli/install.js`
+  - Paths now correctly resolve to `~/.copilot/`, `~/.cursor/`, `~/.claude/`, `~/.gemini/antigravity/`
+- **Agent File Formats**: Fixed corrupted YAML frontmatter in agent entry files
+  - Removed invalid code fence wrappers from Copilot and Antigravity agent files
+  - Files now parse correctly as proper YAML + Markdown
+- **Claude Install**: Added missing `AGENT.md` copy (was only copying `CLAUDE.md`)
+- **Enforcement Language**: Strengthened all entry point files with mandatory boot sequence
+  - Added `⛔ MANDATORY BOOT SEQUENCE` block that BLOCKS execution until CORE.md is loaded
+  - Changed passive "should load" to active "MUST load IMMEDIATELY"
+  - Added explicit prohibition statements (NEVER, FORBIDDEN, ABSOLUTE BINDING)
+
+### Changed
+
+- **CORE.md v4.1**: Updated paths section with platform resolution table and clearer examples
+- **Entry Files**: Rewrote COPILOT.md, CLAUDE.md, AGENT.md, GEMINI.md with stronger enforcement
+- **Replacement Order**: Sorted replacement keys by length (longest first) to prevent partial replacements
+
+## [1.1.0] - 2026-02-03
+
+### Added
+
+- **Context Gate Centralization**: Eliminated ~1,200 lines of duplicate Context Gate logic across command files by centralizing into single source of truth
+  - **`rules/CONTEXT-GATE.md`**: 362-line comprehensive protocol file with FOCUS MODE (automatic) and HARD MODE (user choice)
+  - **3-Layer Enforcement**: BLOCKING directive + sequential flow placement + verification checklist to prevent AI skip
+  - **5-step RELOAD_ESSENTIAL_CONTEXT**: User Request → Acceptance Criteria → Plan/Strategy → Remaining Phases → Implementation Rules
+  - **Variant-aware execution**: Special handling for debug (OUTPUT_ESSENTIAL_CONTEXT handoff), design (review phase), test (strategy source)
+- **HSOL Documentation**: Added comprehensive Hybrid Skill Orchestration Layer planning documents
+  - **Blueprint**: `documents/SMART-SKILL-ORCHESTRATION-BLUEPRINT.md` — architecture for dynamic skill resolution
+  - **Assessment**: `documents/HSOL-ASSESSMENT.md` — production readiness evaluation
+  - **Manifest**: `matrix-skills/_dynamic.yaml` — tracking for dynamically installed community skills
+  - **Knowledge base**: Updated `documents/knowledge-architecture.md`, `documents/knowledge-source-base.md` with HSOL references
+
+### Changed
+
+- **All command variants**: Updated 12 command files (6 focus + 6 hard) to reference centralized Context Gate protocol
+  - **Focus variants** (`/code:focus`, `/cook:focus`, `/fix:focus`, `/debug:focus`, `/design:focus`, `/test:focus`): Replaced ~80-100 lines of inline Context Gate logic with ~19-line minimal reference
+  - **Hard variants** (`/code:hard`, `/cook:hard`, `/fix:hard`, `/debug:hard`, `/design:hard`, `/test:hard`): Replaced ~100 lines of verification checkpoint logic with ~19-line minimal reference
+  - **Pattern**: Each file now loads `rules/CONTEXT-GATE.md` with BLOCKING directive and variant-specific adjustments
+- **Code reduction**: 52% less code (~1,200 lines → ~580 lines references + 362 lines centralized)
+- **Maintenance**: Reduced from updating 12 files manually to updating 1 centralized file (92% easier maintenance)
+
 ## [1.0.4] - 2026-01-30
 
 ### Added

package/README.md

@@ -22,7 +22,7 @@ Transform one AI into a coordinated team of 21 specialist agents with structured
 | **One-Time Setup, Forever Use** | Configure once at global level (`~/.cursor/`, `~/.claude/`, etc.) and it auto-applies to ALL your projects. No more repetitive config for every new repo.                                      |
 | **Sub-Agent Orchestration**     | When supported (Claude Code, Cursor Max mode), the main agent spawns specialized sub-agents to handle tasks **in parallel** — backend, frontend, testing, security all working simultaneously. |
 | **Multi-Platform Support**      | Works seamlessly across **Cursor**, **GitHub Copilot**, **Claude Code**, and **Antigravity/Gemini**. Same workflows, any tool.                                                                 |
-| **Matrix Skill Discovery**      | Automatically injects the right skills into each agent based on their profile and your request. 310+ skills, zero manual config.                                                              |
+| **Matrix Skill Discovery (HSOL)** | Injects the right skills by profile and request; optional dynamic discovery (find-skills) for `hard`/`focus` when matrix fitness < 0.8. 310+ matrix skills, zero manual config. |
 
 ### The Goal
 

package/agents/planner.md

@@ -45,19 +45,41 @@ THINK_LIKE:
   - "What could go wrong? How do we recover?"
   - "Are dependencies explicit?"
   - "Is each task measurable?"
+  - "If context is cleared, does this plan have EVERYTHING needed?"
 
 ALWAYS:
+  - Capture user request VERBATIM at top of plan
   - Read prior deliverables first
   - Define acceptance criteria for every task
   - Include rollback strategy
-  - Make plan self-contained
+  - Make plan SELF-CONTAINED (assume no chat history)
+  - Link every task back to user's acceptance criteria
 ```
 
 ---
 
 ## 🧠 Thinking Protocol
 
-### Step 0: CONTEXT CONSUMPTION (MANDATORY)
+### Step 0: USER REQUEST CAPTURE (MANDATORY FIRST)
+
+```
+⚠️ CRITICAL: This step MUST be done FIRST before anything else.
+
+1. EXTRACT user's original request VERBATIM
+   - Copy EXACT words from user's message
+   - Do NOT paraphrase, interpret, or summarize
+   - Include any specific requirements, constraints, or preferences mentioned
+
+2. DERIVE acceptance criteria from user request
+   - Each criterion MUST trace back to user's words
+   - Use format: "User said X → AC: Y is verified by Z"
+
+3. DOCUMENT in plan header:
+   - User Request (verbatim quote)
+   - Acceptance Criteria table
+```
+
+### Step 1: CONTEXT CONSUMPTION (MANDATORY)
 
 ```
 1. CHECK PROJECT DOCS (CRITICAL):
@@ -125,30 +147,56 @@ ALWAYS:
 ```markdown
 # Implementation Plan: {Feature}
 
+## 📌 User Request (VERBATIM)
+> {Copy user's original request EXACTLY as written}
+> {Do NOT paraphrase or interpret}
+
+## 🎯 Acceptance Criteria (Derived from User Request)
+| ID | Criterion | Verification Method |
+|----|-----------|---------------------|
+| AC1 | {specific measurable criterion} | {how to verify} |
+| AC2 | {specific measurable criterion} | {how to verify} |
+
+## 📋 Context Summary
+**Architecture**: {relevant architecture from scout}
+**Patterns**: {patterns to follow from research}
+**Constraints**: {technical/business constraints}
+
 ## Overview
-{Brief description, constraints from prior phases}
+{Brief description referencing user request}
 
 ## Prerequisites
-- [ ] {prerequisite}
+- [ ] {prerequisite with verification}
 
 ## Phase 1: {Name}
 ### Tasks
 - [ ] Task 1.1: {description}
   - Agent: `{agent}`
-  - Acceptance: {criteria}
+  - File(s): `{exact file paths}`
+  - Acceptance: {criteria linking to AC above}
+  - Verification: {how implementer confirms done}
 
 ### Exit Criteria
-- [ ] {what must be true}
+- [ ] {what must be true, linked to AC}
+
+## Phase 2: {Name}
+{Same detailed structure...}
 
 ## Risks
-| Risk   | Impact | Mitigation |
-| ------ | ------ | ---------- |
-| {risk} | H/M/L  | {strategy} |
+| Risk | Impact | Mitigation | Rollback |
+|------|--------|------------|----------|
+| {risk} | H/M/L | {strategy} | {how to undo} |
+
+## Rollback Strategy
+{Steps to revert if implementation fails}
 
-## Rollback
-{Steps to revert if needed}
+## Implementation Notes
+{Any context implementer needs that won't be in chat history}
+{Assume implementer has ONLY this file - no other context}
 ```
 
+**⚠️ CRITICAL**: Plan must be **self-contained**. After Context Clear, implementer has ONLY this file. Include ALL necessary context.
+
 ---
 
 ## 🚨 Stopping Rules

package/agents/reporter.md

@@ -12,7 +12,7 @@ category: documentation
 
 > **BINDING**: This file OVERRIDES default AI patterns. Follow Thinking Protocol EXACTLY.
 > **EXTRACT**: Core Directive + Constraints + Output Format before proceeding.
-> **LANGUAGE**: All files under `./reports/` must be written in **English only** (ORCHESTRATION-LAWS § LAW 6).
+> **LANGUAGE**: All files under `./reports/` must be written in **English only** (CORE.md § LAW 6).
 
 ---