prompt-cache-skills
Agent skills that find and patch the places where open-source coding agent harnesses break provider prompt caching.
0 members list prompt-cache-skills, unchanged over the last 12 weeks, data as of 16 October 2026
Members who list prompt-cache-skills in their Stack, by week.
Be the first to add prompt-cache-skills
Nobody on Stackness runs prompt-cache-skills yet. Add it to your stack and you're the first member running it.
About prompt-cache-skills
Prompt-cache-skills is a collection of agent skills designed to identify and resolve issues where AI coding agents fail to use provider prompt caching effectively. Developed by OnlyTerp and written in Python, the tool functions as a drop-in compatibility layer for various LLM agent harnesses. It detects incompatibilities between coding agents and prompt caching mechanisms, then applies patches to enable caching where it would otherwise break.
The tool integrates with multiple popular AI coding agents and supports various large language model providers. By pointing an AI coding agent at the repository, developers can automatically equip their agent with the necessary patches to handle prompt caching correctly. This helps optimize performance and reduce costs when using AI agents for code generation and modification tasks.
Checked 16 October 2026.
Nobody on Stackness runs prompt-cache-skills yet.
Sign up and add itPosts about prompt-cache-skills
- Agent context compression tools: Headroom, RTK, Engram, Attic and how each one measured its saving
Seven tools sit between a coding agent and the model as of October 2026. Every headline saving is self-measured, and RTK is the safest place to start.