Build Python data pipelines that liberate structured data from unstructured and semi-structured sources — government PDFs, FOIA releases, scanned reports, scraped HTML, panel-format spreadsheets — into tidy, documented, reproducible civic datasets. Works at six escalating levels, from a one-shot "just give me the CSV"…
Stage 4 of the Clinical ASR Flywheel. Use when priority KER is above 0.3 to run stock NeMo SFT on Parakeet TDT v2 and offline cycle N+1 re-eval. NOT for generic word boosting (use /finetune-asr).
How to phrase tasks and reviews when delegating to Claude Code from Codex via the cc tools. Read before composing a non-trivial ccrescue task or review focus.
Turn any folder of files into a navigable knowledge graph with community detection, an honest audit trail, and three outputs: interactive HTML, GraphRAG-ready JSON, and a plain-language GRAPHREPORT.md.
Scaffolds a new AI Provider plugin for the Drupal AI module. Asks which operation types the provider supports (chat, embeddings, text-to-image, etc.), then generates the plugin class with the correct interface implementations, authentication setup, and model discovery.
Use this skill when the user is working with the OpenMythos codebase — a PyTorch implementation of a hypothesized Recurrent-Depth Transformer (RDT) architecture by Kye Gomez. Trigger on any of these signals: files named main.py in an openmythos/ directory; imports like from openmythos.main import OpenMythos…
Build, debug, and optimize Claude API / Anthropic SDK apps. Apps built with this skill should include prompt caching. TRIGGER when: code imports anthropic/@anthropic-ai/sdk; user asks to use the Claude API, Anthropic SDKs, or Managed Agents (/v1/agents, /v1/sessions, /v1/environments). DO NOT TRIGGER when: code…
Use when fine-tuning a large language model. Covers choosing full vs LoRA/QLoRA, dataset formatting, the transformers/PEFT/TRL stack, key hyperparameters, and evaluating fine-tunes without overfitting.
Complete schema drift repair for the Ghost Contributors NL2SQL system. Patches all four layers (YAML contract, ChromaDB RAG chunks, NL2SQL prompt, RAG prompt) using validated tooling. Includes dry-run preview, step-by-step fallback, validation, and rollback.
A set of rules for running reproducible bioinformatics analyses, which study biological data with software. It uses small self-contained scripts, recorded inputs and outputs, environment details, and version history so results can be repeated.
A five-stage method for planning an enterprise ontology, a structured model of the concepts and relationships used across a business, from source material through ongoing use.
Navigate SAP's AI technology landscape and guide implementation of AI use cases on SAP BTP. Use when someone asks which AI approach to use on SAP, how to build AI apps, agents, or ML models on SAP BTP, or needs guidance on SAP AI Core, Joule, Generative AI Hub, HANA Cloud ML, RPT-1, Document AI, or Joule Skills. Also…
Use when designing or reviewing the tools an LLM agent will call — choosing tool boundaries, argument and result shapes, retrieval/search/get decompositions, MCP or goa-ai toolset surfaces — or when an agent misuses existing tools (wrong tool picked, IDs carried between calls, wrong dates computed, confident answers…
Generic AI-agent skill for installing and wiring the @mobileai/react-native SDK into Expo or React Native apps. Use when a coding assistant needs to add MobileAI, install the React Native AI support agent, configure AIAgent, enable screen mapping, set up voice/dictation/human escalation dependencies, troubleshoot…
File a new source into the LLM Wiki — create its Sources page, append a log entry, apply the proposed cross-reference edits. Use when the user says "ingest this", "file this source", "add this to the wiki", drops a URL/PDF/clipped article, or pastes notes they want captured. Never auto-applies cross-refs without…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: