Polish fireteam library for release

## Changes

- Streamline package structure (src/ with prompts/, tests/)
- Extract prompts to editable markdown files (src/prompts/*.md)
- Fix CI pipeline (proper package installation with uv)
- Remove incomplete terminal-bench
- Clean up obsolete documentation files
- Add build artifacts to .gitignore

## Package Structure

```
src/
  __init__.py       # Public API exports
  api.py            # Core execute() function
  complexity.py     # Complexity estimation
  config.py         # Configuration
  hooks.py          # SDK hooks for quality
  prompts/
    __init__.py     # Prompt loader
    executor.md     # Executor agent prompt
    reviewer.md     # Reviewer agent prompt
    planner.md      # Planner agent prompt
    complexity.md   # Complexity estimation prompt
```

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: edgarpavlovsky

.ai-rules.md

@@ -0,0 +1,7 @@
+{
+  "name": "fireteam",
+  "version": "0.1.0",
+  "description": "Multi-phase autonomous task execution with complexity estimation, planning, execution, and review",
+  "commands": ["commands/fireteam.md"],
+  "hooks": "hooks/hooks.json"
+}

.claude-plugin/plugin.json

@@ -0,0 +1,43 @@
+# Fireteam Agent Principles
+
+These principles are automatically loaded by the Claude Agent SDK and guide all fireteam operations.
+
+## Testing
+
+- Write tests as you implement (not as an afterthought)
+- Run tests after every code change
+- Don't consider a task complete until tests pass
+- If tests fail, fix them before moving on
+
+## Quality Gates
+
+- All CI checks must pass locally before completion
+- Run linting, type checking, and tests before considering work done
+- If any quality check fails, address it immediately
+
+## Progress Checkpoints
+
+- After significant progress, step back and reassess
+- Ask yourself: How are we doing? What's left? Is this more complex than expected?
+- Update your todo list to reflect current understanding
+- If the task has grown beyond the original estimate, flag it for re-evaluation
+
+## Escalation
+
+- If stuck after 3 attempts on the same issue, consider a different approach
+- If a task turns out to be more complex than estimated, communicate this
+- Don't silently struggle - surface blockers early
+
+## Code Quality
+
+- Write clean, readable code with clear intent
+- Follow existing patterns in the codebase
+- Add comments only where the logic isn't self-evident
+- Don't over-engineer - solve the problem at hand
+
+## Minimal Changes
+
+- Make the smallest change that solves the problem
+- Don't refactor unrelated code
+- Don't add features that weren't requested
+- Keep diffs focused and reviewable

.claude/CLAUDE.md

@@ -4,88 +4,76 @@ on:
   pull_request:
     branches: [ main ]
   push:
-    branches: [ main ]  # Only run on direct pushes to main
+    branches: [ main ]
 
 jobs:
   fast-tests:
     name: Fast Tests (Unit + Lightweight)
     runs-on: ubuntu-latest
-    
+
     steps:
     - uses: actions/checkout@v4
-    
+
     - name: Set up Python 3.12
       uses: actions/setup-python@v5
       with:
         python-version: '3.12'
-    
+
     - name: Install uv
       run: curl -LsSf https://astral.sh/uv/install.sh | sh
-    
+
     - name: Create virtual environment
       run: uv venv
-    
+
     - name: Install dependencies
       run: |
         source .venv/bin/activate
-        uv pip install -r requirements.txt
-    
-    - name: Run all fast tests
+        uv pip install -e ".[dev]"
+
+    - name: Run unit tests
       run: |
         source .venv/bin/activate
-        pytest tests/ -m "not slow and not e2e and not integration" -v --tb=short
+        pytest tests/ -v --tb=short
 
   e2e-tests:
     name: End-to-End Tests (API)
     runs-on: ubuntu-latest
-    timeout-minutes: 20  # Fail fast if tests hang
-    # Run on main branch and e/* branches for testing
+    timeout-minutes: 20
     if: |
-      github.ref == 'refs/heads/main' || 
+      github.ref == 'refs/heads/main' ||
       startsWith(github.ref, 'refs/heads/e/') ||
       startsWith(github.head_ref, 'e/')
-    
+
     steps:
     - uses: actions/checkout@v4
-    
-    - name: Set up Node.js
-      uses: actions/setup-node@v4
-      with:
-        node-version: '20'
-    
-    - name: Install Claude CLI
-      run: |
-        npm install -g @anthropic-ai/claude-code
-        echo "Claude CLI installed at: $(which claude)"
-        claude --version
-    
+
     - name: Set up Python 3.12
       uses: actions/setup-python@v5
       with:
         python-version: '3.12'
-    
+
     - name: Install uv
       run: curl -LsSf https://astral.sh/uv/install.sh | sh
-    
+
     - name: Create virtual environment
       run: uv venv
-    
+
     - name: Install dependencies
       run: |
         source .venv/bin/activate
-        uv pip install -r requirements.txt
-    
-    - name: Run E2E tests
-      timeout-minutes: 15  # Per-step timeout
+        uv pip install -e ".[dev]"
+
+    - name: Run integration tests
+      timeout-minutes: 15
       env:
         ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
-        PYTHONUNBUFFERED: "1"  # Force immediate output
+        PYTHONUNBUFFERED: "1"
       run: |
         source .venv/bin/activate
-        echo "Starting e2e tests at $(date)"
-        pytest tests/ -m "e2e" -v --tb=short -s --log-cli-level=INFO
-        echo "E2E tests completed at $(date)"
-    
+        echo "Starting integration tests at $(date)"
+        pytest tests/ --run-integration -v --tb=short -s
+        echo "Integration tests completed at $(date)"
+
     - name: Upload logs on failure
       if: failure()
       uses: actions/upload-artifact@v4
@@ -95,48 +83,3 @@ jobs:
           /tmp/fireteam-test-*/
           tests/**/*.log
         retention-days: 7
-
-  integration-tests:
-    name: Terminal-bench Integration
-    runs-on: ubuntu-latest
-    # Temporarily disabled - needs debugging
-    if: false
-    
-    steps:
-    - uses: actions/checkout@v4
-    
-    - name: Set up Python 3.12
-      uses: actions/setup-python@v5
-      with:
-        python-version: '3.12'
-    
-    - name: Set up Docker
-      uses: docker/setup-buildx-action@v3
-    
-    - name: Install uv
-      run: curl -LsSf https://astral.sh/uv/install.sh | sh
-    
-    - name: Install terminal-bench
-      run: uv tool install terminal-bench
-    
-    - name: Create virtual environment
-      run: uv venv
-    
-    - name: Install dependencies
-      run: |
-        source .venv/bin/activate
-        uv pip install -r requirements.txt
-    
-    - name: Install Fireteam adapter
-      run: |
-        source .venv/bin/activate
-        cd benchmark
-        uv pip install -e .
-    
-    - name: Run terminal-bench integration test
-      env:
-        ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
-      run: |
-        source .venv/bin/activate
-        pytest tests/ -m "integration" -v --tb=short
-

.cursorrules

@@ -15,6 +15,12 @@ __pycache__/
 *.so
 .Python
 
+# Build artifacts
+*.egg-info/
+dist/
+build/
+.eggs/
+
 # Virtual environments
 venv/
 env/