skrun-dev/skrun

ADR Writer generates numbered Architecture Decision Records in the nygard/MADR style. Use it to document an architectural decision, including context, options, decision, consequences, and related ADR links.

Audit Fixture Tool Error invokes an always-failing tool once, then returns a fixed JSON response. Use it to verify tool_call_error event handling in end-to-end tests.

CSV to Executive Report turns operational CSV data into a styled multi-page PDF with narrative sections, charts, and a summary table. Use it when you need a concise executive report or analysis from a CSV file.

Changelog Generator turns local git history or a .git-log.txt dump into a polished CHANGELOG.md and release-notes.md. Use it to summarize commits between tags or draft human-readable release notes.

Code Review analyzes code for bugs, security issues, and code smells, then returns a structured set of findings and a quality score. Use it when you need an audit or actionable improvements.

Data Analyst analyzes CSV or JSON data, finds patterns and outliers, and returns structured insights. Use it for data exploration, reporting, and choosing a chart type for key findings.

Email Drafter writes professional emails from context, tone, and recipient details. Use it to draft business emails with a subject, body, and call to action.

Knowledge Base From Vault turns a folder of Markdown notes into a static HTML knowledge base in a single zip. Use it to publish or share an Obsidian vault, Notion export, or repo docs as a browsable site.

Meeting Transcript To Action Items extracts decisions, action items, and open questions from meeting recordings. It reconciles them with a persistent open-action ledger and writes CSV and recap notes for follow-up tools.

No Manifest Fixture provides a no-manifest fixture for E2E tests where the ecosystem is set to "none". Use it when testing script dependency behavior in a minimal setup.

Node Deps Fixture provides a Node dependencies integration fixture for E2E tests. Use it when you need a scripted test environment for issue #57 coverage.

PDF Processing reads a PDF directly and can extract text, summarize it, or analyze its structure. Use it when a user provides a PDF file and asks for one of those tasks.

PDF Processing extracts text from PDF files, fills forms, and merges documents. Use it when you need to work with PDF content or combine multiple PDFs.

Persistent Context uses persistent context mode for Claude Code workflows. Use it when an agent needs to retain state across steps or sessions.

Python Deps Fixture is used by end-to-end tests to verify the schema, cache, and installer flow for Python dependency handling. It helps confirm the integration chain in script-deps testing.

Receipts To Expenses reads receipt images with vision, classifies each line item into expense categories, and can reconcile them against a bank statement CSV. Use it to generate an Excel workbook and PDF summary for an expense report.

SEO Audit analyzes a website URL for common on-page and technical SEO issues and returns a scored report. Use it to evaluate SEO performance and compare results across audits over time.

Semgrep Rule Creator generates a Semgrep rule bundle from a CVE description and a bad-code example. It is used to draft security rules, tests, and documentation for a specific vulnerability pattern.

Simple Skill is a basic skill for testing agent runtime behavior. Use it when you need a minimal example or a placeholder skill.

Slide Deck Generator turns a Markdown outline into a .pptx deck with title, content, and closing slides. Use it when you need to render talking points or a brief into a presentation file.

Web Scraper browses a page in a headless browser and extracts structured information from its content. Use it when you need factual data from a URL or page sections.