All Skills

Build, run, and validate evaluators for AI/LLM applications with Phoenix, in Python or TypeScript (including Vercel AI SDK apps) — code-based checks and LLM-as-a-judge, run over exported traces or dataframes or as Phoenix experiments, gate them in CI, and validate judges against human-labeled data.

Find out what is going wrong in LLM or agent traffic by reading sampled Phoenix traces, spans, or sessions, writing free-form notes (open coding), then grouping the notes into a few narrow annotations, one per failure dimension with a small label set and counts, that pick eval targets and fix priorities (axial coding). Use for "what's going wrong with this agent", "I just instrumented my app, where do I start", "review these traces", "the chatbot keeps losing context", "what kinds of mistakes is the model making", "what categories of failures do we have", "what should I build evals for", "how do I prioritize fixes", "group these notes", "MECE breakdown" — or any framing that needs observations or categories grounded in real traces rather than invented top-down, even without naming the technique.

OpenInference semantic conventions and instrumentation for Phoenix AI observability. Use when implementing LLM tracing, creating custom spans, or deploying to production.

Store user preferences, learnings from tasks, and procedure outcomes. Use to remember what works and recall context before new tasks. (user)

Store team knowledge, project conventions, and learnings from tasks. Use to remember what works and recall context before new tasks. Connects to Hindsight Cloud. (user)

Expert memory architect. Understands your application, identifies where memory adds value, and produces an implementation plan with bank config, tag schema, and code.

Store team knowledge, project conventions, and learnings from tasks. Use to remember what works and recall context before new tasks. Connects to a self-hosted Hindsight server. (user)

Complete Hindsight documentation for AI agents. Use this to learn about Hindsight architecture, APIs, configuration, and best practices.

Guide for using ToolHive CLI (thv) to run and manage MCP servers, skills, and AI-client plugins. Use when managing `thv ai-plugin` installs, project sync or upgrades, or plugin publishing for Claude Code or Codex; also use for MCP server and skill lifecycle commands. Covers server lifecycle, registries, secrets, client registration, builds, permissions, skills, and AI plugins. NOT for Kubernetes operator usage, ToolHive development/contributing, or general AI-client configuration unrelated to ToolHive.

Query MySQL databases from Codex via MCP

Create a `git` commit.

Use when someone asks for a color palette, a gradient, a background, colors for a brand, site, chart or poster, colors that go with a hex code, or gives a Grabient link. Finds palettes by words or colors, makes them from hex colors, changes one to be lighter, darker, calmer or warmer, shows a set side by side and gives each palette's link, CSS, SVG and PNG. Screen colors only: no print inks, paint, thread or colors taken from a photo.

PR

Activate the understand-pr skill

Activate the gh-bulk-issues skill

Activate the understand-issue skill

Make Moves Issue $1

Commit

Deprecated — activate the understand-pr skill instead

Create New Pull Request

Handle PR Review Comments

Critically review your own PR work before or just after opening the PR.

Deprecated — activate the understand-issue skill instead

Fix GitHub Actions CI Failures