Dependency Analyzer

Dependency Analyzer Skill

Purpose

The Dependency Analyzer skill provides comprehensive dependency analysis capabilities for Python projects. It parses dependency configuration files (requirements.txt, pyproject.toml), builds dependency trees, detects version conflicts, and checks compatibility constraints.

Key Functions:

• Parse and analyze current project dependencies
• Identify new dependencies required for features
• Build visual dependency trees
• Detect version conflicts and incompatibilities
• Check compatibility across Python versions
• Generate structured dependency reports

When to Use

Use this skill when you need to:

• Analyze current project dependencies before adding new features
• Identify all dependencies required for a new feature implementation
• Detect version conflicts between dependencies
• Build dependency trees to understand package relationships
• Verify compatibility of new dependencies with existing ones
• Generate dependency reports for documentation

Typical Scenarios:

• Phase 2 design (dependency analysis for PRP)
• Pre-implementation dependency planning
• Dependency upgrade planning
• Conflict resolution
• Security vulnerability assessment (dependency version analysis)

Workflow

1. Parse Current Dependencies

Action: Read and parse all dependency configuration files in the project

# Check for requirements.txt
if [ -f requirements.txt ]; then
  echo "Found requirements.txt"
  cat requirements.txt
fi

# Check for pyproject.toml
if [ -f pyproject.toml ]; then
  echo "Found pyproject.toml"
  grep -A 30 "dependencies\|requires" pyproject.toml
fi

# Check for setup.py
if [ -f setup.py ]; then
  echo "Found setup.py"
  grep -A 20 "install_requires\|extras_require" setup.py
fi

Output: Inventory of current dependencies with versions

2. Build Dependency Tree

Action: Use dependency-checker script to build complete dependency tree

python .claude/skills/dependency-analyzer/dependency-checker-script.py \
  --analyze \
  --project-root . \
  --output-format tree

Output: Visual dependency tree showing relationships

3. Identify New Dependencies

Action: Extract required libraries from feature analysis

Process:

• Read feature analysis document
• Extract library names from requirements section
• Determine package sources (PyPI, npm, etc.)
• Classify as core vs. development dependencies
• Research latest stable versions

Output: List of new dependencies with recommended versions

4. Check Compatibility

Action: Verify version compatibility and constraints

python .claude/skills/dependency-analyzer/dependency-checker-script.py \
  --check-compatibility \
  --new-deps "package1>=1.0,package2~=2.0" \
  --python-version "3.11"

Checks:

• Python version compatibility (3.11+)
• Dependency version constraints (>=, ~=, ^, ==)
• Breaking changes between versions
• Platform-specific requirements

Output: Compatibility report with warnings and recommendations

5. Detect Conflicts

Action: Identify version conflicts in dependency tree

python .claude/skills/dependency-analyzer/dependency-checker-script.py \
  --check-conflicts \
  --new-deps "package1,package2,package3"

Detects:

• Version conflicts (Package A requires B>=2.0, Package C requires B=0.27.0 | Async HTTP client | core | PyPI |

| pydantic-settings | >=2.0.0 | Settings management | core | PyPI | | pytest-asyncio | >=0.21.0 | Async test support | dev | PyPI |

Dependency Tree

project-root/
├── requests==2.31.0
│   ├── urllib3>=1.21.1,=2017.4.17
│   └── charset-normalizer>=2,=4.6.1
│   └── annotated-types>=0.4.0
└── [new] httpx>=0.27.0
    ├── httpcore>=1.0.0
    ├── certifi
    └── sniffio

Compatibility Check

✅ Python 3.11+ compatible ✅ No breaking changes detected ✅ All version constraints satisfied ⚠️ httpx and requests overlap (both HTTP clients)

Conflicts Detected

None - All dependencies compatible

Recommendations

1. Consider migrating from requests to httpx for async support
2. Pin versions in production: httpx==0.27.2
3. Update requirements.txt with new dependencies
4. Run pip install -r requirements.txt to verify installation

## Best Practices

### Version Specification
- **Use version ranges** for flexibility: `package>=1.0,=1.0` - Minimum version 1.0
  - `~=1.2` - Compatible release (>=1.2, =0.27.0,pydantic-settings>=2.0"

# Detect conflicts
python dependency-checker-script.py --check-conflicts \
  --new-deps "package1,package2"

# Generate full report
python dependency-checker-script.py --full-report \
  --output dependency-report.md

Features:

• Parse requirements.txt, pyproject.toml, setup.py
• Build dependency trees
• Check version compatibility
• Detect conflicts
• Query PyPI for latest versions
• Generate markdown reports

compatibility-matrix.md

Location: .claude/skills/dependency-analyzer/compatibility-matrix.md

Contents:

• Common Python library compatibility rules
• Known incompatible package combinations
• Platform-specific dependency notes
• Python version compatibility matrix
• Resolution strategies for common conflicts

Example Usage

Scenario 1: Pre-Feature Dependency Analysis

Input: Feature requires webhook notifications with async HTTP

Process:

1. Parse current dependencies → Find requests==2.31.0
2. Identify new dependencies → Need httpx for async support
3. Check compatibility → httpx>=0.27.0 compatible with Python 3.11+
4. Detect conflicts → No conflicts (httpx and requests can coexist)
5. Generate report → Structured dependency analysis

Output:

### New Dependencies Required
- httpx>=0.27.0 - Async HTTP client for webhook delivery

### Compatibility: ✅ Compatible
- Python 3.11+ supported
- No conflicts with existing dependencies

### Installation:
```bash
pip install httpx>=0.27.0

Scenario 2: Conflict Detection

Input: Feature requires package-a>=2.0 but existing package-b requires package-a=2.0

1. Detect conflict → Version conflict detected
2. Research solutions → Check if package-b>=2.0 supports package-a>=2.0
3. Generate report → Conflict report with resolution strategy

Output:

### Conflicts Detected
**Conflict**: package-a version conflict
- Feature requires: package-a>=2.0
- package-b==1.5 requires: package-a=2.0)
2. Update requirements.txt:
   - package-a>=2.0
   - package-b>=2.0

**Validation**:
```bash
pip install --dry-run package-a>=2.0 package-b>=2.0

## Integration with Feature Implementation Flow

**Input**: Analysis document with dependencies section

**Process**:
1. Dependency Manager agent activates this skill
2. Skill parses current dependencies from project files
3. Skill extracts new dependencies from analysis document
4. Skill checks compatibility and detects conflicts
5. Skill generates structured dependency report

**Output**: Dependency analysis section for PRP

**Next Step**: Design Orchestrator synthesizes dependency analysis into complete PRP

## Error Handling

### Common Errors

**Missing dependency file**:

Error: No dependency configuration file found Resolution: Create requirements.txt or pyproject.toml

**Invalid version specifier**:

Error: Invalid version specifier "package>=abc" Resolution: Use valid version format (e.g., "package>=1.0.0")

**Conflict detected**:

Warning: Version conflict detected Action: Review conflict report and apply resolution strategy

**PyPI query failure**:

Error: Failed to query PyPI for package "xyz" Resolution: Check network connectivity or package name spelling

## Advanced Features

### Custom Compatibility Rules

Define project-specific compatibility rules in `compatibility-matrix.md`:

```yaml
# Example rule
package-a:
  python: ">=3.11"
  conflicts:
    - package-b=1.0

Dependency Tree Visualization

Generate visual dependency trees for documentation:

python dependency-checker-script.py --analyze --format tree --output tree.txt

Output:

myproject==1.0.0
├── requests==2.31.0
│   ├── urllib3>=1.21.1,=2017.4.17
│   └── charset-normalizer>=2,=4.6.1
│   └── annotated-types>=0.4.0
└── httpx==0.27.2
    ├── httpcore==1.0.2
    ├── certifi
    └── sniffio

---

Version: 2.0.0 Agent: @dependency-manager Phase: 2 (Design & Planning) Created: 2025-10-29

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: matteocervelli
• Source: matteocervelli/llms
• License: MIT

Install and usage instructions live in the source repository linked above.