SkillsAdd

xirothedev/claude-workflow-plugin · Archived

dataset

Search the workflow dataset for best practices from past projects, and help users contribute new run records. Each dataset entry is a real orchestrate-run: domain, stack, architecture, workflows, outcome, lessons. Search it before designing a project so new work starts from accumulated experience. Use when: "search the dataset", "find similar projects", "what worked for X", "contribute to the dataset", "add a dataset entry", "best practices for X".

First seen Jun 24, 2026

Installation

$ npx skills add xirothedev/claude-workflow-plugin --skill dataset
View on skills.sh ↗ GitHub repository ↗

Stronger alternatives

This repository is archived — consider an actively maintained alternative.

langsmith-dataset Similar · more installs
langchain-ai/langsmith-skills

INVOKE THIS SKILL when creating evaluation datasets, uploading datasets to LangSmith, or managi…

4.1K installs
huggingface-datasets Similar · more installs
huggingface/skills

Use this skill for Hugging Face Dataset Viewer API workflows that fetch subset/split metadata, …

1.9K installs
tao-convert-dataset-format Similar · more installs
nvidia/skills

Run `tao-daft convert` to convert NVIDIA TAO DAFT datasets between supported formats. Do not us…

1.6K installs
tao-validate-dataset-format Similar · more installs
nvidia/skills

Run `tao-daft validate` to check NVIDIA TAO DAFT datasets for structure, schema, and cross-refe…

1.5K installs

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

find-skills Same topic
vercel-labs/skills

Helps users discover and install agent skills when they ask questions like "how do I do X", "fi…

3.3M installs
agent-browser Same topic
vercel-labs/agent-browser

Browser automation CLI for AI agents. Use when the user needs to interact with websites, includ…

810.4K installs
vercel-react-best-practices Same topic
vercel-labs/agent-skills 697.8K installs
web-design-guidelines Same topic
vercel-labs/agent-skills

Review UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "chec…

617.3K installs
microsoft-foundry Same topic
microsoft/azure-skills

Build, deploy, evaluate, optimize, fine-tune, and manage Microsoft Foundry agents, models, and …

576.5K installs
azure-prepare Same topic
microsoft/azure-skills

Prepare azd-based Azure projects for deployment: generates azure.yaml, infrastructure (Bicep/Te…

568.3K installs
More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

Stars 2
License MIT
Default branch main
Open issues 0
Status Archived

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 3,010 B
  • docs SUMMARY.md 467 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

Dataset

The plugin ships a dataset of real orchestrate-run records under dataset/. This skill searches it and helps users grow it. The richer the dataset, the better every future orchestrate run starts.

Searching — pull best practices

Two interfaces over the same scoring (dataset/lib.ts):

Via the MCP server (preferred)

The plugin registers the dataset-server MCP server. Call its tools:

  • dataset_search — args: domain, stack[], archetype, keywords[],

limit. Returns ranked entries with their lessons.

  • dataset_get — args: id. Returns one full entry.
  • dataset_stats — no args. Returns entry/domain/archetype counts.

If the MCP server is not connected, fall back to the CLI.

Via the CLI

bun run dataset/search.ts '{"domain":"rest-api","stack":["bun"],"archetype":"verified-swarm","keywords":["auth"],"limit":5}'

When to search

  • Before an architecture interview — search by domain + stack, surface

the architecture and lessons of similar past projects to the user.

  • Before workflow synthesis — search by domain to see which archetype each

phase used and how many rounds it took to converge.

  • When a project hits a pitfall — search by keywords for entries whose

pitfalls describe the same trap.

Always tell the user which entries informed a recommendation — cite the entry id. Treat lessons as evidence, not law: an entry reflects one project.

Contributing — grow the dataset

Run this after a real project ships (orchestrated or hand-built).

  1. Read dataset/README.md and dataset/schema.json.
  2. Draft dataset/entries/<id>.json. Required: id, domain, summary,

stack, architecture, workflows, outcome, lessons, contributor, contributedAt. Be honest about outcome and pitfalls — a failed run with a clear lesson is as valuable as a clean one.

  1. Validate: bun run dataset/validate.ts — must print N/N entries valid.
  2. Open a PR with the new file, or file a Dataset contribution GitHub issue

(.github/ISSUE_TEMPLATE/dataset-contribution.yml) for a maintainer to add.

To help a user contribute: interview them for each required field, write the JSON file, run the validator, and show them the result before opening the PR.

Notes

  • Entry ids are unique and match the file name.
  • The dataset feeds the orchestrate skill — Steps 4 (architecture) and 5

(workflow synthesis) should search it first.