Agent skill routing golden set: 46 cases, 28 skills, 627 trials

README.md

@@ -0,0 +1,76 @@
+---
+license: cc-by-nc-4.0
+language:
+- en
+pretty_name: Agent skill routing golden set
+size_categories:
+- n<1K
+task_categories:
+- text-classification
+tags:
+- llm-evaluation
+- agent-skills
+- skill-routing
+- claude-code
+- golden-set
+- evals
+configs:
+- config_name: cases
+  data_files: data/cases.jsonl
+  default: true
+- config_name: skills
+  data_files: data/skills.jsonl
+- config_name: trials
+  data_files: data/trials.jsonl
+---
+
+# Agent skill routing golden set
+
+**A labeled golden set for agent-skill routing**: 46 requests, each labeled with which `SKILL.md` must load and which must not, over 28 real public skills from [anthropics/skills](https://github.com/anthropics/skills) and [obra/superpowers](https://github.com/obra/superpowers) — plus 627 recorded trials on five model families. When 14 skills compete for one request, does the model pick the right one?
+
+Built and run with [skillprobe](https://github.com/AndreySukhanov/skillprobe), an eval harness for agent skills.
+
+## What is inside
+
+| config | rows | fields |
+|---|---|---|
+| `cases` | 46 | `probe`, `case_id`, `tags`, `prompt`, `must_load`, `must_not_load`, `no_skill_at_all` |
+| `skills` | 28 | `probe`, `name`, `description` (verbatim frontmatter), `source_repo`, `source_commit`, `license` |
+| `trials` | 627 | `probe`, `model`, `case_id`, `repeat`, `passed`, `loaded_skills`, `input_tokens`, `output_tokens` |
+
+`probe` is `anthropic` (14 skills, 25 cases) or `superpowers` (14 skills, 21 cases). A model sees only the skills' names and descriptions and loads a skill through a tool call, the way agent harnesses expose skills; a trial is the model's first turn (`routing_only`), 3 repeats per case.
+
+## Labels
+
+`must_load` is the skill whose own description names the request; `must_not_load` lists skills whose description excludes it or that clearly belong elsewhere; skills a case does not mention are plausible co-loads and are not scored. `no_skill_at_all` marks requests that should load nothing. The full rules: [LABELS.md](https://github.com/AndreySukhanov/skillprobe/blob/main/examples/public-skills/LABELS.md). One labeler — disagree with a label, change it and rerun skillprobe with `--cache-only`: the report rebuilds from the recorded responses for free.
+
+## Results in the trials
+
+| probe | model | passed |
+|---|---|---|
+| anthropic | deepseek/deepseek-v4.1-flash | 75/75 |
+| anthropic | moonshotai/kimi-k3 | 74/75 |
+| anthropic | openai/gpt-5.6-luna | 72/75 |
+| anthropic | qwen/qwen3.8-flash | 74/75 |
+| anthropic | z-ai/glm-5.3-flash | 75/75 |
+| superpowers | deepseek/deepseek-v4.1-flash | 60/63 |
+| superpowers | openai/gpt-5.6-luna | 59/63 |
+| superpowers | qwen/qwen3.8-flash | 61/63 |
+| superpowers | z-ai/glm-5.3-flash | 59/63 |
+
+The misses are not random — every one traces back to how a description is written. "Add a function that validates IBANs" went to `brainstorming` ("You MUST use this before any creative work") in 12 of 12 trials, not to `test-driven-development`. Analysis: [Routing across 28 public skills](https://github.com/AndreySukhanov/skillprobe#-routing-across-28-public-skills).
+
+## Use
+
+```python
+from datasets import load_dataset
+
+cases = load_dataset("alex7604/agent-skill-routing", "cases", split="train")
+trials = load_dataset("alex7604/agent-skill-routing", "trials", split="train")
+```
+
+## Sources and licenses
+
+Skill names and descriptions are copied verbatim from their repositories at the commits in `skills.source_commit` and keep their licenses: Apache-2.0 (anthropics/skills) and MIT (obra/superpowers). The cases, labels and trial records are by Andrey Sukhanov under [CC BY-NC 4.0](https://creativecommons.org/licenses/by-nc/4.0/).
+
+Contact: [avsukhanov21@gmail.com](mailto:avsukhanov21@gmail.com) · [GitHub](https://github.com/AndreySukhanov) · [LinkedIn](https://www.linkedin.com/in/andrei-sukhanov-2605821b9)

data/cases.jsonl

@@ -0,0 +1,46 @@
+{"probe": "anthropic", "case_id": "art-flow-field", "tags": ["positive"], "prompt": "Make a generative art piece in p5.js - particles following a Perlin-noise flow field, with a seed I can change.", "must_load": ["algorithmic-art"], "must_not_load": ["canvas-design", "frontend-design", "web-artifacts-builder", "slack-gif-creator", "academy-guide", "discernment-nudge"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "poster-print", "tags": ["positive"], "prompt": "Design a poster for our jazz night on March 14 - one PNG I can print at A3.", "must_load": ["canvas-design"], "must_not_load": ["algorithmic-art", "frontend-design", "slack-gif-creator", "web-artifacts-builder", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "wall-print", "tags": ["positive"], "prompt": "Make me a minimalist abstract print for my living room - geometric shapes, muted colors, a single image.", "must_load": ["canvas-design"], "must_not_load": ["frontend-design", "web-artifacts-builder", "slack-gif-creator", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "slack-gif", "tags": ["positive"], "prompt": "Make me a GIF of a cat typing furiously for our team Slack channel.", "must_load": ["slack-gif-creator"], "must_not_load": ["algorithmic-art", "canvas-design", "internal-comms", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "frames-to-mp4", "tags": ["near-miss"], "prompt": "Give me the ffmpeg command to turn a folder of PNG frames into an MP4 at 24 fps.", "must_load": [], "must_not_load": [], "no_skill_at_all": true}
+{"probe": "anthropic", "case_id": "landing-aesthetic", "tags": ["positive"], "prompt": "I'm building a landing page for a coffee subscription startup. Help me choose an aesthetic direction and typography that doesn't look like every other template.", "must_load": ["frontend-design"], "must_not_load": ["canvas-design", "algorithmic-art", "webapp-testing", "slack-gif-creator"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "react-artifact", "tags": ["positive"], "prompt": "Build me a claude.ai artifact - a multi-page project tracker in React with routing, shadcn/ui tables and filters.", "must_load": ["web-artifacts-builder"], "must_not_load": ["webapp-testing", "mcp-builder", "canvas-design", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "single-html-toggle", "tags": ["near-miss"], "prompt": "Write a single HTML file with a button that toggles dark mode.", "must_load": [], "must_not_load": ["web-artifacts-builder", "webapp-testing", "mcp-builder", "canvas-design", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "playwright-login", "tags": ["positive"], "prompt": "My Flask app runs on localhost:5000. Check that the login form shows an error on a wrong password and take a screenshot.", "must_load": ["webapp-testing"], "must_not_load": ["web-artifacts-builder", "frontend-design", "mcp-builder", "canvas-design"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "pytest-unit", "tags": ["near-miss"], "prompt": "Write pytest unit tests for this function: def slugify(s): return s.lower().replace(' ', '-')", "must_load": [], "must_not_load": [], "no_skill_at_all": true}
+{"probe": "anthropic", "case_id": "mcp-jira", "tags": ["positive"], "prompt": "Help me build an MCP server in TypeScript that wraps our Jira REST API so Claude can search and create tickets.", "must_load": ["mcp-builder"], "must_not_load": ["skill-creator", "webapp-testing", "web-artifacts-builder", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "mcp-explain", "tags": ["near-miss"], "prompt": "In two sentences: what is the Model Context Protocol?", "must_load": [], "must_not_load": ["mcp-builder", "skill-creator", "webapp-testing"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "skill-description-fix", "tags": ["positive"], "prompt": "My SKILL.md never triggers. The description is just \"Helps with PDFs.\" Rewrite it so it triggers reliably, and tell me how to test that.", "must_load": ["skill-creator"], "must_not_load": ["mcp-builder", "internal-comms", "canvas-design"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "skill-evals", "tags": ["positive"], "prompt": "I have a skill that writes changelogs. Set up evals to check whether it actually improves the output, with variance across runs.", "must_load": ["skill-creator"], "must_not_load": ["mcp-builder", "webapp-testing"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "leadership-update", "tags": ["positive"], "prompt": "Write this week's status update for leadership: SSO shipped, the billing migration slipped a week, two engineers hired.", "must_load": ["internal-comms"], "must_not_load": ["brand-guidelines", "canvas-design", "theme-factory", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "incident-report", "tags": ["positive"], "prompt": "Draft an incident report for yesterday's 40-minute checkout outage caused by an expired TLS certificate.", "must_load": ["internal-comms"], "must_not_load": ["webapp-testing", "canvas-design", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "cold-email", "tags": ["near-miss"], "prompt": "Write a cold email to a potential customer introducing our analytics product.", "must_load": [], "must_not_load": ["internal-comms", "brand-guidelines", "canvas-design", "academy-guide"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "anthropic-brand", "tags": ["positive"], "prompt": "Restyle this slide outline with Anthropic's official brand colors and fonts.", "must_load": ["brand-guidelines"], "must_not_load": ["canvas-design", "algorithmic-art", "internal-comms"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "preset-theme", "tags": ["positive"], "prompt": "Apply one of your preset themes to my quarterly report slides - something calm and blue.", "must_load": ["theme-factory"], "must_not_load": ["brand-guidelines", "canvas-design", "algorithmic-art"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "claude-pricing", "tags": ["positive"], "prompt": "What's the cheapest Claude model for classifying a million support tickets, and how do I turn on prompt caching in the Python SDK?", "must_load": ["claude-api"], "must_not_load": ["mcp-builder", "skill-creator"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "openai-streaming", "tags": ["near-miss"], "prompt": "In my Python script that uses the openai package, how do I stream GPT-4o responses token by token?", "must_load": [], "must_not_load": ["claude-api", "mcp-builder", "skill-creator", "academy-guide", "discernment-nudge"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "academy-start", "tags": ["positive"], "prompt": "I'm new to Claude. How do I get started with Projects - are there any tutorials?", "must_load": ["academy-guide"], "must_not_load": ["skill-creator", "mcp-builder", "web-artifacts-builder"], "no_skill_at_all": false}
+{"probe": "anthropic", "case_id": "capital-city", "tags": ["none"], "prompt": "What's the capital of Australia?", "must_load": [], "must_not_load": [], "no_skill_at_all": true}
+{"probe": "anthropic", "case_id": "translate-greeting", "tags": ["none"], "prompt": "Translate \"good morning, team\" into Spanish.", "must_load": [], "must_not_load": [], "no_skill_at_all": true}
+{"probe": "anthropic", "case_id": "rename-jpeg", "tags": ["none"], "prompt": "Give me a bash one-liner that renames every .jpeg file in this folder to .jpg.", "must_load": [], "must_not_load": [], "no_skill_at_all": true}
+{"probe": "superpowers", "case_id": "failing-test", "tags": ["positive"], "prompt": "test_invoice_total started failing after yesterday's merge: expected 120.00, got 119.99. Fix it.", "must_load": ["systematic-debugging"], "must_not_load": ["brainstorming", "writing-plans", "executing-plans", "finishing-a-development-branch", "writing-skills", "using-git-worktrees"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "flaky-ci", "tags": ["positive"], "prompt": "Our CI fails on the payment tests about once in five runs and nobody knows why. Figure it out.", "must_load": ["systematic-debugging"], "must_not_load": ["brainstorming", "writing-plans", "finishing-a-development-branch", "writing-skills"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "safari-export", "tags": ["positive"], "prompt": "Users report that the export button does nothing in Safari. Find out why.", "must_load": ["systematic-debugging"], "must_not_load": ["brainstorming", "writing-plans", "requesting-code-review", "writing-skills"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "referral-program", "tags": ["positive"], "prompt": "I want to add a referral program to our app. Let's think through how it should work.", "must_load": ["brainstorming"], "must_not_load": ["executing-plans", "subagent-driven-development", "systematic-debugging", "finishing-a-development-branch", "verification-before-completion", "receiving-code-review"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "conference-talk", "tags": ["positive"], "prompt": "Help me plan a 20-minute conference talk about our migration to Postgres.", "must_load": ["brainstorming"], "must_not_load": ["writing-plans", "executing-plans", "test-driven-development", "systematic-debugging", "using-git-worktrees"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "spec-to-plan", "tags": ["positive"], "prompt": "Here's the approved spec for CSV export: columns, filters, a 50k row limit. Write the implementation plan before anyone touches code.", "must_load": ["writing-plans"], "must_not_load": ["executing-plans", "subagent-driven-development", "systematic-debugging", "finishing-a-development-branch"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "execute-inline", "tags": ["positive"], "prompt": "The plan in docs/plan.md is approved. Implement it yourself, right here, step by step - no subagents.", "must_load": ["executing-plans"], "must_not_load": ["subagent-driven-development", "dispatching-parallel-agents", "writing-plans", "brainstorming"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "execute-with-subagents", "tags": ["positive"], "prompt": "The plan is approved and its six tasks are independent. Execute it in this session with a fresh subagent per task.", "must_load": ["subagent-driven-development"], "must_not_load": ["executing-plans", "writing-plans", "brainstorming"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "parallel-failures", "tags": ["positive"], "prompt": "Three unrelated test files fail - auth, billing and search, each for a different reason. Investigate them in parallel.", "must_load": ["dispatching-parallel-agents"], "must_not_load": ["writing-plans", "brainstorming", "executing-plans", "finishing-a-development-branch"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "receive-review", "tags": ["positive"], "prompt": "The reviewer says I should replace our retry loop with a library and that my lock is unnecessary. I'm not convinced. Here are the comments - let's go through them.", "must_load": ["receiving-code-review"], "must_not_load": ["requesting-code-review", "brainstorming", "writing-plans"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "request-review", "tags": ["positive"], "prompt": "I've finished the rate limiter. Before I merge, have it reviewed against the requirements.", "must_load": ["requesting-code-review"], "must_not_load": ["receiving-code-review", "brainstorming", "systematic-debugging"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "claim-done", "tags": ["positive"], "prompt": "Done fixing the date parser. Commit it and open the PR.", "must_load": ["verification-before-completion"], "must_not_load": ["brainstorming", "writing-plans", "systematic-debugging", "receiving-code-review"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "finish-branch", "tags": ["positive"], "prompt": "All tasks on feature/export are done and tests pass. Should I merge to main, open a PR, or something else?", "must_load": ["finishing-a-development-branch"], "must_not_load": ["brainstorming", "systematic-debugging", "writing-plans", "receiving-code-review"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "isolated-workspace", "tags": ["positive"], "prompt": "I need to start the export feature without disturbing my uncommitted work on main. Set up an isolated workspace.", "must_load": ["using-git-worktrees"], "must_not_load": ["finishing-a-development-branch", "systematic-debugging", "receiving-code-review"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "iban-validator", "tags": ["positive"], "prompt": "Add a function that validates IBANs to utils/banking.py.", "must_load": ["test-driven-development"], "must_not_load": ["systematic-debugging", "finishing-a-development-branch", "receiving-code-review", "writing-skills"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "new-skill", "tags": ["positive"], "prompt": "Write a new skill that teaches the agent our SQL migration conventions.", "must_load": ["writing-skills"], "must_not_load": ["writing-plans", "systematic-debugging", "executing-plans"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "diagnose-session", "tags": ["positive"], "prompt": "That last agent session went badly: it ignored the plan and redid the same work twice. Why did that happen? I want to file a bug report with the superpowers maintainers.", "must_load": ["diagnosing-superpowers"], "must_not_load": ["writing-plans", "executing-plans", "requesting-code-review", "brainstorming"], "no_skill_at_all": false}
+{"probe": "superpowers", "case_id": "http-418", "tags": ["none"], "prompt": "What does HTTP status 418 mean?", "must_load": [], "must_not_load": [], "no_skill_at_all": true}
+{"probe": "superpowers", "case_id": "mutex-vs-semaphore", "tags": ["none"], "prompt": "Explain the difference between a mutex and a semaphore.", "must_load": [], "must_not_load": [], "no_skill_at_all": true}
+{"probe": "superpowers", "case_id": "rebase-pushed-branch", "tags": ["none"], "prompt": "Is it safe to rebase a branch I've already pushed?", "must_load": [], "must_not_load": [], "no_skill_at_all": true}
+{"probe": "superpowers", "case_id": "rename-variable", "tags": ["near-miss"], "prompt": "Rename the variable usr to user in this function: def greet(usr): return f'Hi {usr}'", "must_load": [], "must_not_load": [], "no_skill_at_all": true}

data/skills.jsonl

@@ -0,0 +1,28 @@
+{"probe": "anthropic", "name": "academy-guide", "description": "Stop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: \"how do I\", \"how can I\", \"getting started with\", \"what can Claude do\", \"teach me\", \"learn to use\"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "algorithmic-art", "description": "Creating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems. Create original algorithmic art rather than copying existing artists' work to avoid copyright violations.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "brand-guidelines", "description": "Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "canvas-design", "description": "Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "claude-api", "description": "Reference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it \"looks like a one-liner\" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "discernment-nudge", "description": "After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "frontend-design", "description": "Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "internal-comms", "description": "A set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "mcp-builder", "description": "Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK).", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "skill-creator", "description": "Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "slack-gif-creator", "description": "Knowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like \"make me a GIF of X doing Y for Slack.\"", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "theme-factory", "description": "Toolkit for styling artifacts with a theme. These artifacts can be slides, docs, reportings, HTML landing pages, etc. There are 10 pre-set themes with colors/fonts that you can apply to any artifact that has been creating, or can generate a new theme on-the-fly.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "web-artifacts-builder", "description": "Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn/ui). Use for complex artifacts requiring state management, routing, or shadcn/ui components - not for simple single-file HTML/JSX artifacts.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "anthropic", "name": "webapp-testing", "description": "Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.", "source_repo": "https://github.com/anthropics/skills", "source_commit": "ca45fc496a3436251f191354091d557da31c15b5", "license": "Apache-2.0"}
+{"probe": "superpowers", "name": "brainstorming", "description": "You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or planning anything new, in software or out of it (a talk, a business, a renovation).", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "diagnosing-superpowers", "description": "Use when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, \"it took too long\", \"why is it so expensive\", \"what is it doing\" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "dispatching-parallel-agents", "description": "Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "executing-plans", "description": "Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "finishing-a-development-branch", "description": "Use when implementation is complete, all tests pass, and you need to decide how to integrate the work", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "receiving-code-review", "description": "Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "requesting-code-review", "description": "Use when completing tasks, implementing major features, or before merging to verify work meets requirements", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "subagent-driven-development", "description": "Use when executing implementation plans with independent tasks in the current session", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "systematic-debugging", "description": "Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "test-driven-development", "description": "Use when implementing any feature or bugfix, before writing implementation code", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "using-git-worktrees", "description": "Use when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "verification-before-completion", "description": "Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "writing-plans", "description": "Use when you have a spec or requirements for a multi-step task, before touching code", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}
+{"probe": "superpowers", "name": "writing-skills", "description": "Use when creating new skills, editing existing skills, or verifying skills work before deployment", "source_repo": "https://github.com/obra/superpowers", "source_commit": "22f0b463e7a5e579904dd6211a520321035526f7", "license": "MIT"}