Cursor Rules for the llm/ollama Module
These rules are specific to the `llm/ollama` module and supplement `general.cursorrules`. Always adhere to `general.cursorrules` unless explicitly overridden here for a specific reason pertinent to this module.
Explore
11,876 skills indexed with the new KISS metadata standard.
These rules are specific to the `llm/ollama` module and supplement `general.cursorrules`. Always adhere to `general.cursorrules` unless explicitly overridden here for a specific reason pertinent to this module.
You are working on **NeuralDBG**, a causal inference engine for deep learning training dynamics.
These are the general guiding principles for AI-assisted development across the entire Codomyrmex project. Module-specific rules (e.g., in `modules/ai_code_editing.cursorrules`) may supplement or, in rare, justified cases, override these general rules. If a module-specific rule conflicts, it should
"model": "DeepCORO_CLIP",
generic skill
iOS app to translate burned-in video subtitles to user's preferred language.
rule "no-secrets-in-source" {
This is a monorepo for NAV's K9 case handling system (k9-sak-web), containing React/TypeScript frontend applications for case processing. The project includes both the main K9 application and Ungdomsytelse (ung) application.
generic skill
iOS app to translate burned-in video subtitles to user's preferred language.
Please read that file and follow the rules and prompts within it for interacting with this project.
This is a monorepo for NAV's K9 case handling system (k9-sak-web), containing React/TypeScript frontend applications for case processing. The project includes both the main K9 application and Ungdomsytelse (ung) application.
rule "no-secrets-in-source" {
- NEVER run commands automatically (npm install, npm run dev, npm start, etc.)
This is a Figma plugin built with TypeScript, Lit, and Spectrum Web Components. The plugin helps author component option schemas for Adobe Spectrum Design System components.
All dependencies must be peer dependencies.
mini variants) — usually lack depth/context/tool reliability
inputs
fast
including placeholders for iteration or error handling
suggest non-AI or hybrid alternatives (e.g.
practical starter prompt
feedback loops)
bias checks)