Data analysis, machine learning, LLM apps, and prompt design (9.2K skills)
Manage saved prompts in Agent Platform with Python, including creation, listing, retrieval, version tracking, and deletion. Retrieve a specific version and fill template variables to reuse a prompt with its associated model.
Call Gemini and OpenMaaS models such as Llama, DeepSeek, and Qwen through Google Cloud Agent Platform. Prepare authentication and SDK code, run prompts, configure connection endpoints, and troubleshoot common request errors.
Create weekly reports from Postgres metrics, comparing current figures with recent weeks. Add plain-language explanations for notable changes, flag missing data, and post one report to Slack while keeping database access read-only.
Monitor a data warehouse for stale tables, unusual row counts, column changes, and failed loads. Receive alerts with evidence and likely causes, plus GitHub issues for human review, while warehouse data remains unchanged.
Write and debug code for further training large language models with TRL. Match training methods to dataset formats, configure supervised or preference-based training, and set up reward functions through Python or the command line.
Plan and troubleshoot large language and vision model training and inference. This guide uses hardware measurements to identify bottlenecks, address memory shortages and unstable training, and improve checkpointing, recovery, and performance across multiple GPUs and machines.
Collect and update 68 US economic and social indicators, then summarize trends over ten, five, two, and one years. Produce a report covering relationships between indicators, unusual patterns, and topics that warrant further research.
Develop varied ideas for names, writing angles, and problem solving by exploring five candidates internally and presenting the strongest. It also expands small example collections into diverse datasets for evaluation, training, or testing.
Apply over 240 reusable prompt templates to summarize content, extract insights, improve writing, review code, or analyze security threats. It reads each template directly and returns results in the format defined for that task.
Create and improve prompts for agents, workflows, and AI evaluations. Reusable Handlebars templates, which insert data into a fixed structure, help generate consistent instructions from YAML or JSON inputs for use in other tasks.
Review AI agent instructions for unnecessary rules, contradictions, repetition, and vague wording. Get recommendations on what to remove, merge, clarify, or load only when needed, plus an estimate of savings in tokens, the units models process.
Train sentence-transformers models for search, similarity, and reranking, which reorders search results by relevance. Choose a model type, prepare data, configure training, compare results against a baseline, and publish the trained model to Hugging Face Hub.
Configure OpenTelemetry to send AI application traces, records of execution steps, to Confident AI. Reuse existing exporters, choose the regional endpoint, and add AI-specific attributes without requiring the deepeval Python package.
Add DeepEval tracing to Python AI applications so execution steps appear in Confident AI. Choose a supported integration or manual @observe instrumentation, then capture inputs, outputs, step types, tags, and metadata.
Use past local Codex sessions to propose updates to a selected skill and project memory file. Recurring tasks are replayed, changes are checked on separate validation tasks, and proposals are staged for review before adoption.
Review past local Cursor sessions, identify recurring tasks, and propose changes to a project skill. Changes are tested on separate validation tasks and staged for review before adoption, with backups of existing files.
Review past Claude Code sessions and turn recurring tasks into proposed updates to skills and project memory. Replay tasks, compare scores on separate validation tasks, and review staged edits before applying them with backups.
Review past dsh agent sessions to identify recurring tasks and test proposed skill or memory improvements. This skill guides preview runs, validation on reserved tasks, staged proposals, adoption with backups, and scheduled nightly cycles.
Search LangBot Space for plugins, MCP servers, and skills through MCP, a protocol for connecting agents to tools. Inspect item details, check plugin usage compatibility, and review model availability to support setup decisions.
Helps design semantic search and RAG, where an AI answers using retrieved documents. Covers embeddings, numerical representations of content, plus document splitting, search indexes, keyword search combinations, and evaluation of retrieval quality.
Write, debug, and optimize SQL queries for data analysis and database design. The skill covers aggregations, rankings, missing values, and indexes, using execution plans to understand performance and adapting syntax to the database in use.
Adapt large language models to specific tasks using LoRA and QLoRA, methods that train a small set of added parameters. Get guidance on preparing examples, choosing training settings, evaluating outputs, and saving or merging trained adapters.
Plan data ingestion and transformation workflows with Apache Airflow, Spark, and dbt. Get guidance on processing changed records, validating data quality, and rerunning failed or historical jobs without creating duplicate results.
Build and maintain Elasticsearch search and analytics systems with guidance on queries, field types, and aggregations. Plan index migrations, automate data retention, diagnose cluster allocation problems, and investigate slow searches.