tobs-code/SoftPrompt-IR
SoftPrompt-IR is a low-level symbolic annotation layer for LLM prompts, making intent strength, direction, and priority explicit. It is not a DSL or framework, but a minimal, composable way to reduce ambiguity, improve safety, and structure prompts.
When crafting instructions for Large Language Models (LLMs), SoftPrompt-IR helps you explicitly define how important certain aspects of your request are, and whether the model should emphasize or de-emphasize them. It takes your natural language prompts and adds simple, symbolic annotations to make intent strength and direction clear. This is for prompt engineers, content creators, or anyone who frequently interacts with LLMs and needs more predictable outputs.
Use this if you find your LLM prompts are often misinterpreted, produce unexpected results, or need clearer guidance on what information to prioritize or avoid.
Not ideal if you need a full programming language or framework to control LLM behavior, or if you're looking for a tool that forces specific outputs rather than guiding probability.
Stars
11
Forks
—
Language
—
License
Apache-2.0
Category
Last pushed
Feb 11, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/prompt-engineering/tobs-code/SoftPrompt-IR"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
openscilab/memor
Reproducible Structured Memory for LLMs
ronniross/symbiotic-core-library
The Symbiotic Core Library provides a framework of ethical principles, practical modules, and...
ybeven/4D-ARE
Build LLM agents that explain why, not just what. Attribution-driven agent requirements...
Creepybits/World_weaver
The World Weaver System: True AI character consistency using Textual Inheritance. Maintain...
gmasson/systemprompt
A general-purpose system prompt that makes LLMs more rigorous, honest, and genuinely useful,...