observability-langsmith: LLM tracing, evaluation and monitoring skill
A skill by davila7 for LLM observability, covering tracing, evaluation and monitoring of AI applications, released under the MIT license.
Resources for writing, running and reviewing tests, and for checking a build before release.
1,567 results
A skill by davila7 for LLM observability, covering tracing, evaluation and monitoring of AI applications, released under the MIT license.
An agent that plans incremental modernization of legacy systems, covering codebase audits, characterization tests, phased migration and risk control.
An agent that synthesises academic literature into cited reviews with confidence levels, drawing on arXiv and Semantic Scholar.
A command that analyzes a GitHub issue by its number, works on a fix, runs tests and makes a commit.
A SKILL.md file from a community library of cybersecurity skills for AI agents, covering API security testing with Postman.
A Model Context Protocol server that gives LLMs browser automation capabilities through Playwright.
A skill by davila7 that reviews product documents such as PRDs, specs and design briefs for gaps before implementation starts.
A skill that gives guidance on running causal interventions on PyTorch models using pyvene's declarative intervention framework.
An agent that runs a project's test suite, analyses failures and returns a root-cause diagnosis with fix recommendations, without editing code itself.
An agent for TypeScript implementation work that depends on advanced type system patterns, generics and end-to-end type safety.
A command that guides automated git bisect sessions to locate regression commits, using test execution to narrow the search.
An agent definition for finding and removing performance bottlenecks in applications, databases and infrastructure systems.
An agent for Kotlin development covering coroutine patterns, Kotlin Multiplatform code sharing, and Android or server-side work.
A Python agent by davila7 for building type-annotated, async-based code for web APIs, system utilities and larger applications.
An agent for designing and implementing smart contracts, DApps and blockchain protocols, covering Solidity, gas optimization, security auditing and Web3 integration.
A skill for sending designed proteins to a cloud laboratory platform for automated experimental validation.
A skill that gives guidance on using SQLMap to detect and exploit SQL injection vulnerabilities in databases.
An agent that designs user research plans and analyzes user-supplied data to produce design insights.
An agent for building new Rails applications or modernizing existing ones, covering full-stack work with Hotwire and real-time features.
Agent definition for quantitative trading strategy development, derivatives pricing and portfolio risk analytics, published in the claude-code-templates repository.
An agent for building ASP.NET Core web APIs, cloud-native .NET solutions and modern C# applications, covering async patterns, dependency injection and Entity Framework.
A skill that assumes a plan has already failed, then lists the likely causes, ranks them and proposes mitigations.
A skill that compares concrete options by scoring them on weighted criteria and testing their sensitivity and reversibility.
A skill that helps decide whether an AI, LLM, agent or retrieval system needs an evaluation, and how to design and read one.