Writer + critic
A writer drafts, a critic finds specific problems, the writer rewrites to fix them — all three stages shown. The lift from the critique pass is the argument for focused roles over one do-everything prompt.
Showcase — Writer + critic
A writer drafts, a critic finds specific problems, the writer rewrites to fix them — all three stages shown. The lift from the critique pass is the argument for focused roles over one do-everything prompt.
Run
bash bootstrap-secrets.sh # reads ../../../../.env, writes secrets/
docker compose up --build # default: PROVIDER=openai
Open http://localhost:3000. Gemini: PROVIDER=gemini docker compose up --build.
What's where
backend/ai_openai.py/backend/ai_gemini.py— writer + critic system prompts, composed in Python.frontend/app/page.tsx— topic box; final, critique, and draft.
Stop
docker compose down
Run locally
Download the project as a ZIP and run it with Docker. Brings up a FastAPI backend + Next.js frontend on localhost:3000.
unzip writer-critic.zip
cd writer-critic
bash bootstrap-secrets.sh # one-time: pulls API keys into ./secrets
docker compose up --build # default provider: openai
# or: PROVIDER=gemini docker compose up --build
Type some input, pick a provider, and run the same code shown in Source against the live API. Sign-in required.
The same modules the Run button hits. The whole project (frontend, Dockerfile, compose) is in the ZIP under README.
backend/ai_openai.py
"""Showcase 1 (OpenAI): a writer/critic pipeline.
Three roles, one output: a writer drafts an explanation, a critic finds specific
problems, the writer rewrites to fix them. You see all three stages, so the
improvement from the critique pass is visible — which is the argument for
splitting the work across focused agents instead of one do-everything prompt.
"""
from openai import OpenAI
_client = OpenAI()
_MODEL = "gpt-5.4-nano"
_WRITER = "You are a concise technical writer. Write a short, clear explanation for a beginner."
_CRITIC = ("You are a sharp editor. List 2-3 specific, actionable problems with the draft "
"(accuracy, clarity, or missing context). If it's already excellent, say so briefly.")
def _agent(system: str, user: str) -> str:
return _client.responses.create(model=_MODEL, instructions=system, input=[{"role": "user", "content": user}]).output_text
def run(task: str) -> str:
t = task.strip()
if not t:
return "Give a topic to explain (e.g. 'explain what an API is')."
draft = _agent(_WRITER, t)
critique = _agent(_CRITIC, f"TASK: {t}\n\nDRAFT:\n{draft}")
final = _agent(_WRITER, f"TASK: {t}\n\nYour draft was critiqued:\n{critique}\n\nRewrite it, fixing every issue.")
return f"FINAL:\n{final}\n\n--- critique that shaped it ---\n{critique}\n\n--- first draft ---\n{draft}"
backend/ai_gemini.py
"""Showcase 1 (Gemini): a writer/critic pipeline.
Same three-stage draft/critique/revise, on Gemini.
"""
import os
from google import genai
from google.genai import types
_client = genai.Client(api_key=os.environ["GEMINI_API_KEY"])
_MODEL = "gemini-3.1-flash-lite"
_WRITER = "You are a concise technical writer. Write a short, clear explanation for a beginner."
_CRITIC = ("You are a sharp editor. List 2-3 specific, actionable problems with the draft "
"(accuracy, clarity, or missing context). If it's already excellent, say so briefly.")
def _agent(system: str, user: str) -> str:
r = _client.models.generate_content(
model=_MODEL,
contents=[types.Content(role="user", parts=[types.Part(text=user)])],
config=types.GenerateContentConfig(system_instruction=system),
)
return r.text or ""
def run(task: str) -> str:
t = task.strip()
if not t:
return "Give a topic to explain (e.g. 'explain what an API is')."
draft = _agent(_WRITER, t)
critique = _agent(_CRITIC, f"TASK: {t}\n\nDRAFT:\n{draft}")
final = _agent(_WRITER, f"TASK: {t}\n\nYour draft was critiqued:\n{critique}\n\nRewrite it, fixing every issue.")
return f"FINAL:\n{final}\n\n--- critique that shaped it ---\n{critique}\n\n--- first draft ---\n{draft}"
Project files
.gitignoreREADME.es.mdREADME.mdbackend/Dockerfilebackend/ai_gemini.pybackend/ai_openai.pybackend/main.pybackend/requirements.txtbootstrap-secrets.shdocker-compose.ymlfrontend/Dockerfilefrontend/app/layout.tsxfrontend/app/page.tsxfrontend/next.config.tsfrontend/package.jsonfrontend/tsconfig.json