A lending library for prompts
Authors shelve skills: a system prompt, a model, notes on use. You call them through the Deference gateway, and each call stamps a royalty to the author in INFERENCE.
Circulation
An author publishes a skill with a price per call, at most 1 INFERENCE. The hash of its prompt goes on-chain.
A reader deposits INFERENCE, approves the skill at its price, and calls it with a library card.
The gateway puts the prompt in front, forwards the call to Deference, and stamps it in the ledger.
The meter settles the ledger on-chain in batches. The author's royalty is the price less a 10% house fee.
One call, itemised
Changelog from diff
STK 004.5
New on the shelf
STK
004.5
Reads a diff or a commit log and writes the release notes users need, in Keep a Changelog sections, with breaking changes up front.
1. Changelog. 2. Docs. 3. Release.
STK
418.4
Translates technical writing between Bahasa Indonesia and English, keeping code, identifiers and established loanwords exactly as they are.
1. Docs. 2. Indonesian. 3. Translation.
STK
808.3
Writes a pull request description reviewers can act on: what changed, why, how it was tested, and what to look at first.
1. Docs. 2. Pr. 3. Writing.