Trusted skills for AI agents from official vendors and established maintainers. Every skill is attributed to its author, scanned, and comes with install steps for your AI tool.
284 skills
DatadogOfficial
Automate evaluation by fitting an LLM-judge to your existing human-labelled annotation queues.
DatadogOfficial
Classify user satisfaction in Datadog Agent Observability traces and sessions using trace data and RUM behavioral signals.
RedisOfficial
Implement semantic caching for LLM responses using Redis LangCache to reduce latency and API costs.
PydanticOfficial
Run offline AI evaluations and review results directly in Logfire.
PydanticOfficial
Build production-grade AI agents using the Pydantic AI framework.
FirecrawlOfficial
Search and analyze a specialized corpus of biomedical, life-science, and technical research papers.
ReplicateOfficial
A framework for evaluating and comparing Replicate models based on cost, performance, and capabilities.
ReplicateOfficial
A guide for discovering and evaluating AI models hosted on the Replicate platform.
PostHogOfficial
Analyze agent intent clusters and tool usage patterns in PostHog MCP to optimize tool discoverability and performance.
NeonOfficial
A unified API gateway for accessing multiple LLMs and embedding models directly within your Neon database branches.
NeonOfficial
Implement Managed Better Auth for Neon-hosted applications to handle user identity, sessions, and OAuth.
WeaviateOfficial
A comprehensive index of architectural patterns and blueprints for building AI applications with Weaviate.
WeaviateOfficial
Integrate Weaviate Engram to provide persistent, long-term memory for LLM agents and chatbots.
PostHogOfficial
Integrate PostHog data, third-party connectors, and interactive analytics into custom canvases using the ph SDK.
PostHogOfficial
Integrate PostHog error tracking into Next.js applications for monitoring and issue management.
PostHogOfficial
Automated hygiene scout for the PostHog skills store that audits agent skills for authoring violations.
PostHogOfficial
A guide for interpreting and managing PostHog project access control rules using MCP tools.
PostHogOfficial
Provisions an experiment-specific Replay Vision scanner to analyze user behavior across experiment variants.
PostHogOfficial
A systematic workflow for qualifying and engaging PostHog customers approaching billing transitions.
PostHogOfficial
Integrate PostHog error tracking into Python applications for automated exception monitoring and issue management.
PostHogOfficial
Integrate PostHog task and canvas comments into your agent workflow using the PostHog MCP exec dispatcher.
PostHogOfficial
Analyze, compare, and report on offline AI evaluation experiments and scorer history.
PostHogOfficial
Identifies and ranks the most relevant session recordings for a specific error tracking issue.
PostHogOfficial
An autonomous PostHog scout that identifies cross-product correlations and monitors surfaces not covered by specialist agents.