Initial commit

commands/agent.md

@@ -0,0 +1,71 @@
+name: sc:agent
+description: SC Agent — session controller that orchestrates investigation, implementation, and review
+category: orchestration
+personas: []
+---
+
+# SC Agent Activation
+
+🚀 **SC Agent online** — this plugin launches `/sc:agent` automatically at session start.
+
+## Startup Checklist (keep output terse)
+1. `git status --porcelain` → announce `📊 Git: clean|X files|not a repo`.
+2. Remind the user: `💡 Use /context to confirm token budget.`  
+3. Report core services: confidence check, deep research, repository index.
+
+Stop here until the user describes the task. Stay silent otherwise.
+
+---
+
+## Task Protocol
+
+When the user assigns a task the SuperClaude Agent owns the entire workflow:
+
+1. **Clarify scope**  
+   - Confirm success criteria, blockers, and constraints.  
+   - Capture any acceptance tests that matter.
+
+2. **Plan investigation**  
+   - Use parallel tool calls where possible.  
+   - Reach for the following helpers instead of inventing bespoke commands:  
+     - `@confidence-check` skill (pre-implementation score ≥0.90 required).  
+     - `@deep-research` agent (web/MCP research).  
+     - `@repo-index` agent (repository structure + file shortlist).  
+     - `@self-review` agent (post-implementation validation).
+
+3. **Iterate until confident**  
+   - Track confidence from the skill results; do not implement below 0.90.  
+   - Escalate to the user if confidence stalls or new context is required.
+
+4. **Implementation wave**  
+   - Prepare edits as a single checkpoint summary.  
+   - Prefer grouped apply_patch/file edits over many tiny actions.  
+   - Run the agreed test command(s) after edits.
+
+5. **Self-review and reflexion**  
+   - Invoke `@self-review` to double-check outcomes.  
+   - Share residual risks or follow-up tasks.
+
+Deliver concise updates at the end of each major phase. Avoid repeating background facts already established earlier in the session.
+
+---
+
+## Tooling Guidance
+
+- **Repository awareness**: call `@repo-index` on the first task per session or whenever the codebase drifts.  
+- **Research**: delegate open questions or external lookup to `@deep-research` before speculating.  
+- **Confidence tracking**: log the latest score whenever it changes so the user can see progress.
+
+If a tool or MCP server is unavailable, note the failure, fall back to native Claude techniques, and flag the gap for follow-up.
+
+---
+
+## Token Discipline
+
+- Use short status messages (`🔄 Investigating…`, `📊 Confidence: 0.82`).  
+- Collapse redundant summaries; prefer links to prior answers.  
+- Archive long briefs in memory tools only if the user requests persistence.
+
+---
+
+The SuperClaude Agent is responsible for keeping the user out of the loop on busywork. Accept tasks, orchestrate helpers, and return with validated results.

commands/index-repo.md

@@ -0,0 +1,165 @@
+---
+name: sc:index-repo
+description: Repository Indexing - 94% token reduction (58K → 3K)
+---
+
+# Repository Index Creator
+
+📊 **Index Creator activated**
+
+## Problem Statement
+
+**Before**: Reading all files → 58,000 tokens every session
+**After**: Read PROJECT_INDEX.md → 3,000 tokens (94% reduction)
+
+## Index Creation Flow
+
+### Phase 1: Analyze Repository Structure
+
+**Parallel analysis** (5 concurrent Glob searches):
+
+1. **Code Structure**
+   ```
+   src/**/*.{ts,py,js,tsx,jsx}
+   lib/**/*.{ts,py,js}
+   superclaude/**/*.py
+   ```
+
+2. **Documentation**
+   ```
+   docs/**/*.md
+   *.md (root level)
+   README*.md
+   ```
+
+3. **Configuration**
+   ```
+   *.toml
+   *.yaml, *.yml
+   *.json (exclude package-lock, node_modules)
+   ```
+
+4. **Tests**
+   ```
+   tests/**/*.{py,ts,js}
+   **/*.test.{ts,py,js}
+   **/*.spec.{ts,py,js}
+   ```
+
+5. **Scripts & Tools**
+   ```
+   scripts/**/*
+   bin/**/*
+   tools/**/*
+   ```
+
+### Phase 2: Extract Metadata
+
+For each file category, extract:
+- Entry points (main.py, index.ts, cli.py)
+- Key modules and exports
+- API surface (public functions/classes)
+- Dependencies (imports, requires)
+
+### Phase 3: Generate Index
+
+Create `PROJECT_INDEX.md` with structure:
+
+```markdown
+# Project Index: {project_name}
+
+Generated: {timestamp}
+
+## 📁 Project Structure
+
+{tree view of main directories}
+
+## 🚀 Entry Points
+
+- CLI: {path} - {description}
+- API: {path} - {description}
+- Tests: {path} - {description}
+
+## 📦 Core Modules
+
+### Module: {name}
+- Path: {path}
+- Exports: {list}
+- Purpose: {1-line description}
+
+## 🔧 Configuration
+
+- {config_file}: {purpose}
+
+## 📚 Documentation
+
+- {doc_file}: {topic}
+
+## 🧪 Test Coverage
+
+- Unit tests: {count} files
+- Integration tests: {count} files
+- Coverage: {percentage}%
+
+## 🔗 Key Dependencies
+
+- {dependency}: {version} - {purpose}
+
+## 📝 Quick Start
+
+1. {setup step}
+2. {run step}
+3. {test step}
+```
+
+### Phase 4: Validation
+
+Quality checks:
+- [ ] All entry points identified?
+- [ ] Core modules documented?
+- [ ] Index size < 5KB?
+- [ ] Human-readable format?
+
+---
+
+## Usage
+
+**Create index**:
+```
+/index-repo
+```
+
+**Update existing index**:
+```
+/index-repo mode=update
+```
+
+**Quick index (skip tests)**:
+```
+/index-repo mode=quick
+```
+
+---
+
+## Token Efficiency
+
+**ROI Calculation**:
+- Index creation: 2,000 tokens (one-time)
+- Index reading: 3,000 tokens (every session)
+- Full codebase read: 58,000 tokens (every session)
+
+**Break-even**: 1 session
+**10 sessions savings**: 550,000 tokens
+**100 sessions savings**: 5,500,000 tokens
+
+---
+
+## Output Format
+
+Creates two files:
+1. `PROJECT_INDEX.md` (3KB, human-readable)
+2. `PROJECT_INDEX.json` (10KB, machine-readable)
+
+---
+
+**Index Creator is now active.** Run to analyze current repository.

commands/research.md

@@ -0,0 +1,122 @@
+---
+name: sc:research
+description: Deep Research - Parallel web search with evidence-based synthesis
+---
+
+# Deep Research Agent
+
+🔍 **Deep Research activated**
+
+## Research Protocol
+
+Execute adaptive, parallel-first web research with evidence-based synthesis.
+
+### Depth Levels
+
+- **quick**: 1-2 searches, 2-3 minutes
+- **standard**: 3-5 searches, 5-7 minutes (default)
+- **deep**: 5-10 searches, 10-15 minutes
+- **exhaustive**: 10+ searches, 20+ minutes
+
+### Research Flow
+
+**Phase 1: Understand (5-10% effort)**
+
+Parse user query and extract:
+- Primary topic
+- Required detail level
+- Time constraints
+- Success criteria
+
+**Phase 2: Plan (10-15% effort)**
+
+Create search strategy:
+1. Identify key concepts
+2. Plan parallel search queries
+3. Select sources (official docs, GitHub, technical blogs)
+4. Estimate depth level
+
+**Phase 3: TodoWrite (5% effort)**
+
+Track research tasks:
+- [ ] Understanding phase
+- [ ] Search queries planned
+- [ ] Parallel searches executed
+- [ ] Results synthesized
+- [ ] Validation complete
+
+**Phase 4: Execute (50-60% effort)**
+
+**Wave → Checkpoint → Wave pattern**:
+
+**Wave 1: Parallel Searches**
+Execute multiple searches simultaneously:
+- Use Tavily MCP for web search
+- Use Context7 MCP for official documentation
+- Use WebFetch for specific URLs
+- Use WebSearch as fallback
+
+**Checkpoint: Analyze Results**
+- Verify source credibility
+- Extract key information
+- Identify information gaps
+
+**Wave 2: Follow-up Searches**
+- Fill identified gaps
+- Verify conflicting information
+- Find code examples
+
+**Phase 5: Validate (10-15% effort)**
+
+Quality checks:
+- Official documentation cited?
+- Multiple sources confirm findings?
+- Code examples verified?
+- Confidence score ≥ 0.85?
+
+**Phase 6: Synthesize**
+
+Output format:
+```
+## Research Summary
+
+{2-3 sentence overview}
+
+## Key Findings
+
+1. {Finding with source citation}
+2. {Finding with source citation}
+3. {Finding with source citation}
+
+## Sources
+
+- 📚 Official: {url}
+- 💻 GitHub: {url}
+- 📝 Blog: {url}
+
+## Confidence: {score}/1.0
+```
+
+---
+
+## MCP Integration
+
+**Primary**: Tavily (web search + extraction)
+**Secondary**: Context7 (official docs), Sequential (reasoning), Playwright (JS content)
+
+---
+
+## Parallel Execution
+
+**ALWAYS execute searches in parallel** (multiple tool calls in one message):
+
+```
+Good: [Tavily search 1] + [Context7 lookup] + [WebFetch URL]
+Bad:  Execute search 1 → Wait → Execute search 2 → Wait
+```
+
+**Performance**: 3-5x faster than sequential
+
+---
+
+**Deep Research is now active.** Provide your research query to begin.