upamune/radicaster

Overview

This skill is a TAILWIND documentation assistant that helps locate and summarize information from the project's docs. It searches Markdown files in the docs/ directory and returns concise, citation-backed answers. It is optimized for quick lookups and accurate references.

How this skill works

The skill inspects Markdown files stored under docs/ and uses a search utility to find relevant sections. It extracts frontmatter metadata (source_url and fetched_at) and includes those fields in every response so users can verify and follow up. A provided search script supports JSON output and result limits for programmatic use.

When to use it

• When you need a concise explanation of TAILWIND features documented in the repo.
• When you must locate configuration examples, API details, or usage notes from the docs/ directory.
• When preparing documentation-based answers that require source citations and fetch dates.
• When automating lookups via the search script for tooling or integrations.
• When confirming whether documentation has changed since the fetched_at date.

Best practices

• Always cite the source_url and include the fetched_at date in the response.
• Prefer quoting short, relevant excerpts and then summarizing in plain language.
• Use the search script (python scripts/search_docs.py) with --json for structured results.
• Limit search results with --max-results when you only need top matches.
• State uncertainty if the fetched_at date suggests docs may have changed since retrieval.

Example use cases

• Summarize the documented steps for configuring a TAILWIND feature and provide the source link.
• Find a code example from docs/ and paste a minimal excerpt with attribution.
• Automate validation by running the search script and parsing JSON output for CI checks.
• Answer a user question about behavior that is described in the documentation, including fetch date.