I write status reports every Friday, and it is always the same busywork. Today we will wire up a small Python CLI that turns raw bullet points into a polished weekly update. It calls Oxlo.ai directly, and because Oxlo.ai charges per request instead of per token, I do not have to trim my commit history to save money. See https://oxlo.ai/pricing for current plan details.
What you'll need
- Python 3.10 or newer
- The OpenAI SDK:
pip install openai - An Oxlo.ai API key from https://portal.oxlo.ai
Step 1: Configure the Oxlo.ai client
Oxlo.ai exposes an OpenAI-compatible endpoint, so the official SDK drops right in. Replace the placeholder with your actual key before running the script.
from openai import OpenAI
client = OpenAI(base_url="https://api.oxlo.ai/v1", api_key="YOUR_OXLO_API_KEY")Step 2: Define the system prompt
The system prompt locks the tone and structure. I treat this as the contract between my code and the model, and you can edit the sections to match your team's format.
SYSTEM_PROMPT = """You are a technical writing assistant that converts raw engineering bullet points into a concise weekly status report.
Rules:
- Output valid markdown.
- Include sections: Accomplishments, Blockers, Goals for Next Week.
- Keep each bullet to one or two sentences.
- Use a calm, professional tone.
- Do not invent facts that are not in the input."""Step 3: Format the user input
I read bullet points from a text file and wrap them in a simple message. This keeps the CLI clean and lets me pipe data from other tools.
def load_bullets(path: str) -> str:
with open(path, "r", encoding="utf-8") as f:
return f.read().strip()
def build_user_message(bullets: str) -> str:
return f"Turn the following bullet points into a weekly status report.\n\nBullet points:\n{bullets}"Step 4: Stream the report
We call Llama 3.3 70B, Oxlo.ai's general-purpose flagship. Streaming lets me watch the report appear in real time instead of waiting for the full response.
from openai import OpenAI
client = OpenAI(base_url="https://api.oxlo.ai/v1", api_key="YOUR_OXLO_API_KEY")
SYSTEM_PROMPT = """You are a technical writing assistant that converts raw engineering bullet points into a concise weekly status report.
Rules:
- Output valid markdown.
- Include sections: Accomplishments, Blockers, Goals for Next Week.
- Keep each bullet to one or two sentences.
- Use a calm, professional tone.
- Do not invent facts that are not in the input."""
def build_user_message(bullets: str) -> str:
return f"Turn the following bullet points into a weekly status report.\n\nBullet points:\n{bullets}"
def generate_report(bullets: str) -> None:
user_message = build_user_message(bullets)
response = client.chat.completions.create(
model="llama-3.3-70b",
messages=[
{"role": "system", "content": SYSTEM_PROMPT},
{"role": "user", "content": user_message},
],
stream=True,
temperature=0.4,
)
print("\n--- Generated Report ---\n")
for chunk in response:
if chunk.choices[0].delta.content:
print(chunk.choices[0].delta.content, end="")
print()Step 5: Wire up the CLI
Argparse gives us a proper command-line interface. I also added a --model flag so you can try Qwen 3 32B or DeepSeek V3.2 on Oxlo.ai without changing code.
import argparse
from openai import OpenAI
client = OpenAI(base_url="https://api.oxlo.ai/v1", api_key="YOUR_OXLO_API_KEY")
SYSTEM_PROMPT = """You are a technical writing assistant that converts raw engineering bullet points into a concise weekly status report.
Rules:
- Output valid markdown.
- Include sections: Accomplishments, Blockers, Goals for Next Week.
- Keep each bullet to one or two sentences.
- Use a calm, professional tone.
- Do not invent facts that are not in the input."""
def load_bullets(path: str) -> str:
with open(path, "r", encoding="utf-8") as f:
return f.read().strip()
def build_user_message(bullets: str) -> str:
return f"Turn the following bullet points into a weekly status report.\n\nBullet points:\n{bullets}"
def main():
parser = argparse.ArgumentParser(description="Generate a weekly status report via Oxlo.ai")
parser.add_argument("--input", required=True, help="Path to a text file with bullet points")
parser.add_argument("--model", default="llama-3.3-70b", help="Oxlo.ai model ID")
args = parser.parse_args()
bullets = load_bullets(args.input)
user_message = build_user_message(bullets)
response = client.chat.completions.create(
model=args.model,
messages=[
{"role": "system", "content": SYSTEM_PROMPT},
{"role": "user", "content": user_message},
],
stream=True,
temperature=0.4,
)
print("\n--- Generated Report ---\n")
for chunk in response:
if chunk.choices[0].delta.content:
print(chunk.choices[0].delta.content, end="")
print()
if __name__ == "__main__":
main()Run it
Save your bullet points to bullets.txt:
- Fixed race condition in auth service causing intermittent 500s
- Migrated staging RDS instance to Graviton 4
- Drafted RFC for new caching layerThen run the script:
python status_report.py --input bullets.txt --model llama-3.3-70bYou should see streamed markdown like this:
--- Generated Report ---
## Accomplishments
- Fixed a race condition in the auth service that was causing intermittent 500 errors.
- Migrated the staging RDS instance to Graviton 4 processors.
## Blockers
- None noted.
## Goals for Next Week
- Finalize the RFC for the new caching layer and circulate it for review.Next steps
Add JSON mode to the Oxlo.ai call so you can pipe the structured sections straight into a Notion or Slack dashboard. If your team writes mostly code, swap in DeepSeek V3.2 on Oxlo.ai for tighter technical language. Both models sit on the same endpoint, so it is a one-line change.
