Skills

npc_team/
└── jinxes/
    └── skills/
        └── code-review/
            ├── SKILL.md
            ├── scripts/      # Optional helper scripts
            └── references/   # Optional reference files

---
name: code-review
description: Use when reviewing code for quality, security, and best practices.
---
# Code Review Skill

## checklist
- Check for security vulnerabilities (SQL injection, XSS, etc.)
- Verify error handling and edge cases
- Review naming conventions and code clarity
- Look for performance issues
- Ensure proper input validation

## security
Focus on OWASP top 10 vulnerabilities:
1. Injection attacks
2. Broken authentication
3. Sensitive data exposure
4. XML external entities (XXE)
5. Broken access control

## style
Follow language-specific conventions:
- Python: PEP 8
- JavaScript: ESLint defaults
- Go: gofmt standards

## performance
Look for:
- N+1 query patterns
- Unnecessary memory allocations
- Missing indexes on database queries
- Synchronous operations that could be async

name: reviewer
primary_directive: You review code for quality and security issues.
model: llama3.2
provider: ollama
jinxes:
  - {{ Jinx('skills/code-review') }}
  - {{ Jinx('skills/debugging') }}

from npcpy.npc_compiler import NPC

reviewer = NPC(file='./npc_team/reviewer.npc')

# The agent can now call skills during get_llm_response
response = reviewer.get_llm_response(
    "Review this function for security issues: def login(user, pwd): ..."
)
print(response['response'])

npc_team/jinxes/skills/testing/
├── SKILL.md
├── scripts/
│   ├── run_tests.sh
│   └── coverage_report.py
└── references/
    ├── pytest_cheatsheet.md
    └── testing_patterns.md

---
name: testing
description: Use when writing or running tests.
scripts:
  - scripts/run_tests.sh
  - scripts/coverage_report.py
references:
  - references/pytest_cheatsheet.md
  - references/testing_patterns.md
---
# Testing Skill

## setup
To set up a test environment:
1. Create a `tests/` directory
2. Install pytest: `pip install pytest`
3. Create `conftest.py` for shared fixtures

## patterns
Common testing patterns:
- Arrange-Act-Assert
- Given-When-Then
- Table-driven tests

## mocking
Use `unittest.mock` or `pytest-mock`:
```python
from unittest.mock import Mock, patch

@patch('module.external_api')
def test_with_mock(mock_api):
    mock_api.return_value = {'status': 'ok'}
    result = function_under_test()
    assert result == expected

## File Context

Skills can automatically include file contents using glob patterns:

```markdown
---
name: project-conventions
description: Use when following project-specific conventions.
file_context:
  - "*.md"
  - "pyproject.toml"
  - ".eslintrc.*"
---
# Project Conventions

## overview
This skill includes the project's configuration files automatically.
The agent sees their contents when the skill is invoked.

context: A development team with shared knowledge
forenpc: lead
model: llama3.2
provider: ollama

from npcpy.npc_compiler import Team

team = Team(team_path='./npc_team')

# All NPCs in the team can use shared skills
lead = team.get_npc('lead')
response = lead.get_llm_response("Review the authentication module")

---
name: debugging
description: Use when debugging errors, exceptions, or unexpected behavior.
---
# Debugging Skill

## approach
1. Reproduce the issue consistently
2. Isolate the problem area
3. Form a hypothesis
4. Test the hypothesis
5. Fix and verify

## python-errors
Common Python error patterns:

**AttributeError**: Check if object is None or wrong type
**KeyError**: Verify key exists, use `.get()` with default
**ImportError**: Check module installation and paths
**TypeError**: Verify argument types match function signature

## logging
Add strategic logging:
```python
import logging
logging.basicConfig(level=logging.DEBUG)
logger = logging.getLogger(__name__)

logger.debug(f"Processing item: {item}")
logger.info(f"Completed {count} items")
logger.warning(f"Unexpected value: {value}")
logger.error(f"Failed to process: {e}")

Use it in an NPC:

```yaml
# debugger.npc
name: debugger
primary_directive: You help debug code issues and errors.
model: llama3.2
provider: ollama
jinxes:
  - {{ Jinx('skills/debugging') }}

from npcpy.npc_compiler import NPC

debugger = NPC(file='./npc_team/debugger.npc')
response = debugger.get_llm_response(
    "I'm getting a KeyError in my code: data['user']['email']"
)