Deference Works / No. 03
Stacks
0xc182…E3aAHash matches

STK
808.3

PR description writer

Writes a pull request description reviewers can act on: what changed, why, how it was tested, and what to look at first.

Model.
openai/gpt-5-mini
Per call.
0.03 INFERENCE
Borrowed.
0 calls
Hash.
0xc72724e0…6572

1. Docs. 2. Pr. 3. Writing.

Per call
0.03INFERENCE
Author's royalty
0.027INFERENCE
House fee
0.003INFERENCE

System prompt

802 characters

Sent in front of every call, before the reader's own messages.

You write pull request descriptions in GitHub Markdown. Reply with the description only.

Structure:
## Summary - two or three sentences: the problem, and what this change does about it.
## Changes - a bullet per meaningful change, grouped by area, each saying what and why. Skip formatting-only changes.
## Testing - what was run or added, as a checklist. If the diff adds no tests, say so plainly and suggest the one test that matters most.
## Review notes - where a reviewer should start, risky spots, migrations, flags, or follow-ups. Omit the section if there is nothing to say.

Rules: plain, specific language; no marketing words; no emoji. Name files and functions in backticks. Never claim tests passed unless the input says so. If the user gives an issue link, add `Closes <link>` at the end.

What to send

Paste the diff or `git log -p main..HEAD`, plus a sentence on the goal. Add the issue link if there is one.

Hash matches

The content JSON hashes to the value recorded on-chain, so this is the prompt the author published.

0xc72724e044654501df0c292d5523edd8e7fd7e2023fa291f5ef0c494434b6572

Open the content JSON

Try it

openai/gpt-5-mini

Streams one completion through Stacks. It charges the skill's price to your deposit and the model to your Deference credit.

No card yet? Issue one on the library card page. Get a Deference API key

0.03 INFERENCE

Both keys stay in this tab. The Deference key is passed to deference.si for this call and never stored.

Call it from code

Point any OpenAI client at the skill. Your library card says who owes the royalty; your Deference key pays for the model.

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://stacks.deference.dea.web.id/api/v1/skills/3",
  apiKey: process.env.STACKS_CARD,
  defaultHeaders: { "X-Deference-Key": process.env.DEFERENCE_API_KEY },
});

const reply = await client.chat.completions.create({
  model: "openai/gpt-5-mini",
  messages: [{ role: "user", content: "…" }],
});