Skill UI

This guide provides production-ready code patterns for integrating with the Cohere AI API v2 using TypeScript and Python. It covers essential best practices, including singleton client management, robust retry mechanisms, type-safe chat wrappers, streaming, embedding, reranking, and structured JSON output. Use this when establishing team coding standards or refactoring existing Cohere integrations.

This skill transforms a course syllabus (PDF, DOCX, etc.) into a highly curated, supplementary academic reading list. It parses topics and learning outcomes, calibrates content difficulty based on the target audience (undergraduate, graduate, etc.), and searches academic databases (Consensus) for recent, peer-reviewed papers. The output is a professional .docx file containing summaries, links, and advanced discussion questions, making it invaluable for educators and students needing structured, up-to-date course resources.

Open Dynamic Workflows (ODW) is an advanced, open-source engine designed to manage complex coding tasks using multiple AI agents. It operates in three stages: planning the task into parallel subtasks, orchestrating various coding agents (like Codex or Antigravity) to work concurrently, and finally, performing an adversarial verification pass to challenge and validate the combined output before merging. It supports multiple model providers and is ideal for structured, high-reliability software development.

This framework guides AI agents to critically assess their own output across five key axes: Accuracy, Completeness, Clarity, Actionability, and Conciseness. It provides a structured, evidence-based scorecard and concrete suggestions for improvement, turning every task completion into a deep learning opportunity.

OpenSquilla-compatible adapter that generates browser-playable AI audio such as voiceovers, narration, music, and sound effects through OpenRouter audio models. It streams OpenAI gpt-audio output, writes WAV assets, and reports structured AUDIO_READY manifests for webpage integration.

This skill parses complex, free-text clinical trial eligibility criteria from sources like ClinicalTrials.gov into structured, machine-readable rules. It then evaluates these criteria against a comprehensive patient profile (derived from OpenMed data) to determine potential eligibility. The output provides an explainable verdict (eligible, ineligible, or unknown) for every criterion, serving as a critical decision support tool for clinicians and researchers.

This skill manages complex, stateful proof projects as a local-first pipeline. It guides the user through drafting, rigorous verification, and source bookkeeping. It systematically handles proof continuation across runs, maintaining a full audit trail. When local attempts stall, it orchestrates handoffs to external AI models (like GPT Pro) or performs optional deep audits (DeepSeek), ensuring a structured, traceable, and verifiable output suitable for academic or technical submission.

This skill processes raw mentions of conditions and diagnoses extracted from clinical notes. It performs synonym deduplication, filters out negated or hypothetical mentions, and assigns a structured clinical status (active, resolved, or historical). The output is a clean, reconciled problem list, structured for USCDI Problem or FHIR Condition elements.

Restructures any vague build, research, or legal-drafting request into a highly structured IRAC (Issue, Rule, Analysis, Conclusion) prompt. This method optimizes output quality for frontier models by forcing the definition of constraints, underlying rules, required reasoning, and precise deliverables, moving beyond simple prompting.

This tool serves as a pre-flight safety check for legal prompts before submission to third-party AI surfaces (e.g., ChatGPT, Claude). It assesses whether the content risks compromising attorney-client privilege or work-product protection. The output provides a structured safety band (SAFE/CAUTION/STOP), detailed triggered risk factors, a discovery impact summary, and a legally safe, redacted rewrite.

Build persistent AI agents on Azure AI Foundry using the Microsoft Agent Framework Python SDK, with function tools, hosted tools, streaming, threads, and structured outputs.

This skill enables structured AI red team exercises to test for jailbreak resistance, data exfiltration risks, and harmful output controls. It helps discover exploitable failure modes before attackers do, suitable for new LLM features, third-party model evaluation, and security audits. Includes Python scripts for automated testing against various attack vectors like prompt injection and tool abuse.