ralphmode
Configure Claude Code, Codex CLI, and Gemini CLI for Ralph-style automation with fewer approval prompts while keeping project boundaries, secret denylists, and…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 11-05 00:16
Agents & models
25,653 results
permission-max
Reduce repeated permission prompts across Microsoft Scout, Copilot CLI, and host tool confirmations with user-approved settings and explicit before/after…
retro-workspace
Reflect reusable learnings into the current workspace / repository design and automation assets (.github/**, AGENTS.md, repo scripts/tasks). Not for ~/.copilot…
find-skills
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express…
mcp-builder
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when…
skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's…
verify-cascade
Verification-gated model escalation — run a cheap model, verify its output with decomposed per-field checks, and escalate to a strong model only when a flag…
wrap-up
Close out a launched Claude Managed Agent — recap every primitive the founder now owns, regenerate the single-file overview page, and suggest the next 1-2…
alibabacloud-wuying-console
Query and manage Alibaba Cloud WUYING resources from a local agent through the official Aliyun CLI product plugins. Use for cloud computers, cloud browsers,…
dnn-tuning
[DNN tuning] Use when the user says "DNN tuning", "deep learning tuning", "adjust network architecture", "adjust dropout", "DNN overfitting". Based on Optuna TPE Bayesian optimization, searches network architecture (number of layers, width, dropout) and training parameters (learning_rate, weight_decay, batch_size), with diagnostic-driven constrained search space. Prerequisite…
feature-analysis
[Feature Deep Analysis] Use when the user needs to "comprehensively analyze features", "generate a feature report", "check correlations", "recommend modeling plans", "run through the features", "feature panorama", or "which features are useful". Includes four major dimensions: basic statistics, IV value, PSI stability, and correlation analysis, and outputs four modeling feature plans. Prerequisite: the target variable must be provided. Does not do single-feature quick exploration, does not do model training. With...
insurance-agent-application-check
A professional skill that provides insurance agents with pre-application material completeness checks and application form filling assistance, helping agents self-check before submission to avoid policy returns due to missing materials or filling errors. Triggered when agents need to check whether materials are complete, whether the application form is filled in correctly, or whether health disclosures are omitted while helping clients apply for insurance. Suitable for scenarios such as pre-application self-checks, standardized submission processes, and standardized operation guidance for new agents.…
insurance-agent-customer-profiling
Intelligent 360-degree customer profiling and analysis tool for insurance agents. Based on information such as customer policy data, claims records, and interaction behavior, it automatically generates a dynamic tagging system, protection gap matrix, lifecycle positioning, and business development strategy recommendations. Use when users mention customer profiling, customer analysis, customer tags, protection gaps, business development strategy, customer 360, customer insights, or policy analysis. Trigger words: customer profiling, customer situation, profile update.
insurance-agent-objection-handler
A professional skill that provides insurance agents with client objection handling scripts and strategies, helping agents effectively respond to common reasons for refusal, resolve doubts, and advance toward closing. It is triggered when agents encounter objections such as the premium being too expensive, not needed, need to think it over, or no time, or when they face silence or stalling after presenting a plan. It applies to scenarios such as initial resistance during the needs analysis stage, price objections after plan presentation, and handling final doubts during the closing stage. Trigger words: too expensive, think it over, not needed.
insurance-agent-product-wiki
Insurance agent product knowledge base and sales assistant. Upload product documents to automatically ingest them, and build an independent knowledge page for each product based on the LLM Wiki concept. Integrates three major modes: learning Q&A, simulated drills, and script recommendations. Use this skill when the user mentions insurance products, product ingestion, script recommendations, simulated drills, learning products, or insurance Q&A. Trigger words: what does the product cover, script drills, difference between critical illness and medical.
insurance-agent-renewal-assist
Professional skills for insurance agents to provide policy renewal reminders, renewal maintenance scripts, and customer churn prevention strategies, helping agents improve customer renewal rates and reduce policy lapse caused by overdue premiums. Triggered when a customer's policy is about to expire and needs renewal, when the customer has delayed paying renewal premiums, or when a pre-renewal policy health check is needed. Suitable for scenarios such as reminders 30 days before renewal, dunning for overdue payments, retention when the customer is considering surrender, and using the renewal occasion to increase coverage.…
lr-tuning
[LR Scorecard Tuning] Use when the user says "LR tuning", "scorecard optimization", "adjust the number of bins", "adjust the C value", or "LR overfitting". Based on Optuna TPE Bayesian optimization, jointly search WoE binning parameters (max_n_bins, iv_threshold) and LR model parameters (C, regularization), with diagnosis-driven constrained search space. Prerequisites: requires first using...
model-comparison
[Multi-model performance comparison] Use when the user needs to "compare algorithms", "multi-model comparison", "compare XGB and LR", "try all three algorithms", "which algorithm is better", "run them all and compare", "compare after automatic hyperparameter tuning", or "fair comparison". Based on the same data split, run multiple algorithms such as XGBoost / LR scorecard / DNN side by side, fill in multi-dimensional metrics such as Brier/BCR/true PSI, and do…
xgb-modeling
[XGBoost Modeling] Use when the user needs to "train a model", "run a baseline", "build a model to try", "compare the effects of feature schemes", "make a prediction", "run an xgb", "look at AUC/KS/BCR". Supports multi-feature scheme comparison, AUC/KS/BCR/Brier evaluation, Bootstrap confidence intervals, probability calibration curves, stability analysis (Mann-Kendall trend test). Prerequisites…
xgb-tuning
[XGBoost Hyperparameter Tuning — The Only Entry Point for Tuning] Use when the user says "help me tune parameters", "what should I do if the model is overfitting", "adjust parameters such as learning_rate/max_depth". Core capabilities: based on Optuna TPE Bayesian optimization +…
elicit-llm-creativity
Produce distinctive creative work by widening ideation, injecting external randomness, grounding choices in taste, and using independent critique. Use for…
context-compressor
Maximize signal-per-token in agent context windows. Use whenever context is being assembled for a task and the question is "what to include, what to cut, and…
planning
State machine orchestrator. Mutable discovery, PRD construction, and full PLAN→EXECUTE→EMIT→VERIFY→COMPLETE lifecycle. Invoke at session start and on any new…
magi
Put a hard question to three independent LLM CLIs from different vendors and return a tallied verdict (decision mode) or a cross-checked research…