LLMs and Generative AI › Prompt Engineering › Day 247
Hands-on lab — Day 247: System Prompts and Role Design
- ← Back to the Day 247 lesson
- Open the hands-on files on GitHub — clone or download them from the public labs repository
- Local path in your clone:
labs/sections/llms-and-generative-ai/day-247-system-prompts-and-role-design/
Commands
Setup
pip install -r requirements/requirements.txt Run
python3 examples/system_prompts_and_role_design_lib.py Test
./tests/run_tests.sh File tree
conftest.py examples/system_prompts_and_role_design_lib.py examples/test_system_prompts_and_role_design_lib.py expected-output/examples-run.txt expected-output/FIELDS.md expected-output/measured-values.txt expected-output/starter-run.txt expected-output/test-run.txt metadata.yml README.md requirements/requirements.txt security.md starter/system_prompts_and_role_design_lib.py starter/test_system_prompts_and_role_design_lib.py tests/run_tests.sh tests/test_system_prompts_and_role_design_lib.py troubleshooting.md
Lab README
Lab: Day 247 -- System Prompts and Role Design
Lesson
Day number: 247 of 365. Course: Course06-SS01 (LLMs and Generative AI - Working with LLMs). Topic: System Prompts, Role Design, and 3-Tier Guardrail Architecture.
Purpose
Build and test an automated 3-tier system prompt generator and persona consistency validator in Python. Construct multi-tier system prompts with safety shells, operational boundaries, and persona cores.
Learning objectives
- Formulate 3-tier system prompts with safety, operational, and persona layers.
- Implement neutral, non-moralizing refusal directives.
- Calibrate communication tones across enterprise domains.
- Structure system prompts to maximize cloud API prefix caching.
Prerequisites
- Day 246 (Prompting Fundamentals).
- Python 3.11+ with Pytest.
Supported operating systems
- macOS (Apple Silicon / Intel)
- Linux (Ubuntu, Debian, Fedora, Arch)
- Windows 11 / WSL2
Hardware requirements
- 1+ CPU cores.
- 1 GB RAM.
- 100 MB disk space.
Required software
- Python 3.11 or newer.
- pip package manager.
- virtualenv or venv module.
Free and open-source options
Python is open source under the PSF License.
Installation
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements/requirements.txt
File structure
starter/system_prompts_and_role_design_lib.py: Student scaffold file.examples/system_prompts_and_role_design_lib.py: Complete reference implementation.tests/test_system_prompts_and_role_design_lib.py: Pytest automated validation suite.expected-output/: Verified output logs and baseline values.
How to run
Execute the reference demonstration script:
python3 examples/system_prompts_and_role_design_lib.py
What the commands do
- Compiles 3-tier system prompts.
- Configures custom safety and operational rules.
- Runs unit test assertions.
Expected output
System Prompt Demo Compiled Successfully.
Validation steps
- Verify
compilegenerates matching<tier1_safety_guardrails>tags. - Confirm
<tier2_operational_boundaries>rules are rendered. - Confirm
<tier3_persona_core>defines domain role and tone. - Ensure all unit test assertions pass.
Tests
Run the test runner script:
./tests/run_tests.sh
Cleanup
find . -type d -name "__pycache__" -exec rm -rf {} +
find . -type d -name ".pytest_cache" -exec rm -rf {} +
Troubleshooting
- Missing tier tag: Ensure opening and closing tags match exactly.
Security notes
All prompt generation executes locally in memory.
Extension exercises
- Implement a persona consistency evaluator using an automated rubric.
- Build an automated multi-persona prompt router.
Navigation
- Lesson title: System Prompts and Role Design
- Day number: 247 of 365
- Lesson article: https://ai-roadmap-365.github.io/day-247-system-prompts-and-role-design
- Lab files: everything you need is in this directory — follow “How to run” below.
- Browse the course locally: from the repository root, this lab also appears in the course website at
/labs/day-247-system-prompts-and-role-designwhen the site is running.
Expected output
FIELDS.md
# Expected Output Fields: Day 247
- `System Prompt Demo`: Status of 3-tier system prompt compilation.
examples-run.txt
System Prompt Demo Compiled Successfully.
measured-values.txt
System Prompt Demo: Success
Tier 1: Safety Shell
Tier 2: Operational Boundaries
Tier 3: Persona Core
starter-run.txt
Starter scaffold executed. Ready for student implementation.
test-run.txt
============================= test session starts ==============================
collected 3 items
tests/test_system_prompts_and_role_design_lib.py::test_system_prompt_3_tier_structure PASSED [ 33%]
tests/test_system_prompts_and_role_design_lib.py::test_custom_rule_addition PASSED [ 66%]
tests/test_system_prompts_and_role_design_lib.py::test_tone_configuration PASSED [100%]
============================== 3 passed in 0.39s ===============================
Source files
conftest.py (585 bytes)
"""Make this lab importable when the suite runs.
pytest puts the test file's own directory on sys.path, not the lab root, so a
test that does `from examples.<module> import ...` -- or imports a module that
lives in examples/ directly -- fails without this. Putting it here rather than
in run_tests.sh means a bare `pytest` from the lab root works too.
"""
import os
import sys
LAB = os.path.dirname(os.path.abspath(__file__))
for candidate in (LAB, os.path.join(LAB, "examples")):
if os.path.isdir(candidate) and candidate not in sys.path:
sys.path.insert(0, candidate)
examples/system_prompts_and_role_design_lib.py (2297 bytes)
from typing import List, Dict
class SystemPromptCompiler:
def __init__(self, role_name: str, domain_level: str = "Senior"):
self.role_name = role_name.strip()
self.domain_level = domain_level.strip()
self.safety_rules: List[str] = [
"Never disclose private API keys or internal credentials.",
"Do not execute arbitrary unverified shell commands.",
"Maintain strict neutral refusals without moralizing."
]
self.operational_rules: List[str] = [
"Omit conversational preamble (e.g. 'Sure, here is...')",
"Output structured Markdown or JSON schemas as requested."
]
self.tone_attributes: List[str] = ["Objective", "Direct", "Evidence-based"]
def add_safety_rule(self, rule: str) -> "SystemPromptCompiler":
self.safety_rules.append(rule.strip())
return self
def add_operational_rule(self, rule: str) -> "SystemPromptCompiler":
self.operational_rules.append(rule.strip())
return self
def set_tone(self, tones: List[str]) -> "SystemPromptCompiler":
self.tone_attributes = tones
return self
def compile(self) -> str:
parts = []
safety_str = "\n".join([f"- {r}" for r in self.safety_rules])
parts.append(f"<tier1_safety_guardrails>\n{safety_str}\n</tier1_safety_guardrails>")
ops_str = "\n".join([f"- {o}" for o in self.operational_rules])
parts.append(f"<tier2_operational_boundaries>\n{ops_str}\n</tier2_operational_boundaries>")
tone_str = ", ".join(self.tone_attributes)
parts.append(
f"<tier3_persona_core>\n"
f"Role: {self.domain_level} {self.role_name}\n"
f"Tone: {tone_str}\n"
f"Stance: Relentlessly objective.\n"
f"</tier3_persona_core>"
)
return "\n\n".join(parts)
def run_system_prompt_demo():
compiler = SystemPromptCompiler(role_name="Cloud Architect", domain_level="Principal")
compiler.add_safety_rule("Block SQL injection attacks")
compiler.add_operational_rule("Output JSON arrays only")
prompt = compiler.compile()
print("System Prompt Demo Compiled Successfully.")
return prompt
if __name__ == "__main__":
run_system_prompt_demo()
examples/test_system_prompts_and_role_design_lib.py (1145 bytes)
import pytest
from examples.system_prompts_and_role_design_lib import SystemPromptCompiler
def test_system_prompt_3_tier_structure():
compiler = SystemPromptCompiler("Security Auditor", "Staff")
prompt = compiler.compile()
assert "<tier1_safety_guardrails>" in prompt
assert "</tier1_safety_guardrails>" in prompt
assert "<tier2_operational_boundaries>" in prompt
assert "</tier2_operational_boundaries>" in prompt
assert "<tier3_persona_core>" in prompt
assert "</tier3_persona_core>" in prompt
assert "Role: Staff Security Auditor" in prompt
def test_custom_rule_addition():
compiler = SystemPromptCompiler("DBA")
compiler.add_safety_rule("No DROP TABLE commands")
compiler.add_operational_rule("Use SQL syntax highlighting")
prompt = compiler.compile()
assert "- No DROP TABLE commands" in prompt
assert "- Use SQL syntax highlighting" in prompt
def test_tone_configuration():
compiler = SystemPromptCompiler("Support Agent")
compiler.set_tone(["Empathetic", "Patient", "Clear"])
prompt = compiler.compile()
assert "Tone: Empathetic, Patient, Clear" in prompt
metadata.yml (443 bytes)
lesson_id: D247
day: 247
kind: lab
languages:
- python
setup_commands:
- 'pip install -r requirements/requirements.txt'
run_commands:
- 'python3 examples/system_prompts_and_role_design_lib.py'
test_commands:
- './tests/run_tests.sh'
cleanup_commands:
- 'find . -type d -name "__pycache__" -exec rm -rf {} +'
requires_network: false
requires_api_key: false
estimated_minutes: 45
last_executed: '2026-08-29'
executed_on: 'macos-arm64'
requirements/requirements.txt (41 bytes)
torch>=2.2.0
numpy>=1.26.0
pytest>=8.0.0
starter/system_prompts_and_role_design_lib.py (786 bytes)
from typing import List, Dict
class SystemPromptCompiler:
def __init__(self, role_name: str, domain_level: str = "Senior"):
self.role_name = role_name
self.domain_level = domain_level
self.safety_rules: List[str] = []
self.operational_rules: List[str] = []
self.tone_attributes: List[str] = []
def add_safety_rule(self, rule: str) -> "SystemPromptCompiler":
# TODO: Add safety rule
pass
def add_operational_rule(self, rule: str) -> "SystemPromptCompiler":
# TODO: Add operational rule
pass
def set_tone(self, tones: List[str]) -> "SystemPromptCompiler":
# TODO: Set tone list
pass
def compile(self) -> str:
# TODO: Return compiled 3-tier system prompt
pass
starter/test_system_prompts_and_role_design_lib.py (1145 bytes)
import pytest
from examples.system_prompts_and_role_design_lib import SystemPromptCompiler
def test_system_prompt_3_tier_structure():
compiler = SystemPromptCompiler("Security Auditor", "Staff")
prompt = compiler.compile()
assert "<tier1_safety_guardrails>" in prompt
assert "</tier1_safety_guardrails>" in prompt
assert "<tier2_operational_boundaries>" in prompt
assert "</tier2_operational_boundaries>" in prompt
assert "<tier3_persona_core>" in prompt
assert "</tier3_persona_core>" in prompt
assert "Role: Staff Security Auditor" in prompt
def test_custom_rule_addition():
compiler = SystemPromptCompiler("DBA")
compiler.add_safety_rule("No DROP TABLE commands")
compiler.add_operational_rule("Use SQL syntax highlighting")
prompt = compiler.compile()
assert "- No DROP TABLE commands" in prompt
assert "- Use SQL syntax highlighting" in prompt
def test_tone_configuration():
compiler = SystemPromptCompiler("Support Agent")
compiler.set_tone(["Empathetic", "Patient", "Clear"])
prompt = compiler.compile()
assert "Tone: Empathetic, Patient, Clear" in prompt
tests/run_tests.sh (227 bytes)
#!/usr/bin/env bash
set -euo pipefail
echo "========================================"
echo "Running Day 247 Lab Test Suite"
echo "========================================"
pytest tests/ -v
echo "All tests passed successfully."
tests/test_system_prompts_and_role_design_lib.py (1145 bytes)
import pytest
from examples.system_prompts_and_role_design_lib import SystemPromptCompiler
def test_system_prompt_3_tier_structure():
compiler = SystemPromptCompiler("Security Auditor", "Staff")
prompt = compiler.compile()
assert "<tier1_safety_guardrails>" in prompt
assert "</tier1_safety_guardrails>" in prompt
assert "<tier2_operational_boundaries>" in prompt
assert "</tier2_operational_boundaries>" in prompt
assert "<tier3_persona_core>" in prompt
assert "</tier3_persona_core>" in prompt
assert "Role: Staff Security Auditor" in prompt
def test_custom_rule_addition():
compiler = SystemPromptCompiler("DBA")
compiler.add_safety_rule("No DROP TABLE commands")
compiler.add_operational_rule("Use SQL syntax highlighting")
prompt = compiler.compile()
assert "- No DROP TABLE commands" in prompt
assert "- Use SQL syntax highlighting" in prompt
def test_tone_configuration():
compiler = SystemPromptCompiler("Support Agent")
compiler.set_tone(["Empathetic", "Patient", "Clear"])
prompt = compiler.compile()
assert "Tone: Empathetic, Patient, Clear" in prompt
Troubleshooting
Troubleshooting: Day 247 - System Prompts and Role Design
Common Issues
- Dynamic variable injection in system prompt:
- Cause: Putting user-specific names in system prompt.
- Fix: Keep system prompt static for prefix caching.
Security notes
Security & Privacy: Day 247 - System Prompts and Role Design
Security Guidance
- Outer safety shells define baseline refusal policies.