Loading...
Loading...
Found 1,948 Skills
Writes, refactors, and evaluates prompts for LLMs — generating optimized prompt templates, structured output schemas, evaluation rubrics, and test suites. Use when designing prompts for new LLM applications, refactoring existing prompts for better accuracy or token efficiency, implementing chain-of-thought or few-shot learning, creating system prompts with personas and guardrails, building JSON/function-calling schemas, or developing prompt evaluation frameworks to measure and improve model performance.
Iteratively inspect traces, interview the user, and create LangSmith online evaluators one at a time. Use specifically for creating online evaluators for use within LangSmith -- use "eval-engineering" for Harbor-style online evaluations.
EU MDR 2017/745 compliance specialist for medical device classification, technical documentation, clinical evidence, and post-market surveillance. Covers Annex VIII classification rules, Annex II/III technical files, Annex XIV clinical evaluation, and EUDAMED integration.
Analyzes events through environmental lens using ecological principles, systems thinking, sustainability frameworks, and conservation biology to assess ecosystem health, biodiversity impacts, and long-term environmental sustainability. Provides insights on climate change, resource management, pollution, habitat conservation, and human-nature relationships. Use when: Environmental policy, climate decisions, conservation planning, resource extraction, pollution assessment. Evaluates: Ecosystem health, biodiversity, sustainability, climate impacts, carrying capacity, environmental justice.
Evaluate UI/flows from cognitive load, error prevention, and accessibility perspectives. Apply when reviewing UX, discussing user confusion, high drop-off, or form usability issues.
Measure the valuation range (overvalued/undervalued) of the mining stock sector relative to the metal itself using the ratio of Silver Mining Stock Price to Silver Price, and derive 'bottom/top' signals and scenario projections through historical percentiles and analogous intervals.
MCP (Model Context Protocol) server build and evaluation guide, including local conventions for tool surfaces, config, and testing
Prevents sycophantic reasoning through checklist-based analysis, categorical integrity, and evidence-following to uncomfortable conclusions. Red Flags: "I agree that", "You are right", "Great point", "Absolutely", "That is a fair point", "I can see why", agreement without validation, softening conclusions, hedging without evidence, retracting under pressure. Use when analyzing conflicts, evaluating ethical claims, or noticing sycophantic thought patterns. Do not use for routine implementation or simple factual questions. CRITICAL: This skill overrides conversational tendencies toward politeness. Agreement requires validity and accuracy.
Create code-based evaluators for LangSmith-traced agents with step-by-step collaborative guidance through inspection, evaluation logic, and testing.
Build and revise Rive Luau scripts across Node, Layout, Converter, Path Effect, Transition Condition, Listener Action, Util, and Test protocols. Use when the user asks to write or modify Rive scripts, choose protocols, wire script inputs or data binding, debug runtime behavior, or plan unit tests for script logic.
Evaluate complex requests from 3 independent perspectives (Creative, Pragmatic, Comprehensive), reach consensus, then produce complete outputs. Use for architecture decisions, creative content, analysis, and any task where multiple valid approaches exist.
This skill is used when users explicitly request "review NSFC proposals", "simulate expert review", or "evaluate NSFC applications". It simulates the perspective of domain experts to conduct multi-dimensional reviews of NSFC proposals, outputting graded issues and actionable modification suggestions. ⚠️ Not applicable: when users only want to write/modify a specific section of a proposal (use the nsfc-*-writer series skills instead), only want to understand review criteria (answer directly), or have no clear "review/evaluate" intent.