oaustegard

Analyze and categorize Bluesky accounts by topic using keyword extraction. Use when users mention Bluesky account analysis, following/follower lists, topic discovery, account curation, or network analysis.

Convenes expert panels for problem-solving. Use when the user mentions panel, experts, multiple perspectives, MECE, DMAIC, RAPID, Six Sigma, root cause analysis, strategic decisions, or process improvement.

Generates guardrailed UI from natural language. Claude emits constrained JSON, which the skill runtime renders via Preact.

Creates production-ready MCP servers using FastMCP v2. Use when building MCP servers, optimizing tool descriptions for context efficiency, implementing progressive disclosure for multiple capabilities, or packaging servers for distribution.

Provides ranked content search over any text corpus using BM25. It's corpus-agnostic, working across cloned repos, knowledge stores, uploaded files, and local directories, building an in-memory index each time for multi-word content search beyond grep.

Securely manages API credentials for multiple providers (Anthropic Claude, Google Gemini, GitHub). Use when skills need to access stored API keys for external service invocations.

Validates development tool installations across Python, Node.js, Java, Go, Rust, C/C++, Git, and system utilities. Use when verifying environments or troubleshooting dependencies.

Delivers a static Hello World HTML demo page with a bookmarklet. Use when the user requests the hello demo, hello world demo, or demo page.

Exports project instructions and knowledge files from the current Claude project. Use when users want to clone, copy, backup, or export a project's configuration and files.

Build and cache a personalized container environment from a Dockerfile-like spec, supporting both single-layer and multi-layer composition with independent layer caching.

Creates browser-executable JavaScript bookmarklets with strict formatting requirements. Use when users mention bookmarklets, browser utilities, dragging code to bookmarks bar, or need JavaScript that runs when clicked in the browser toolbar.

Retrieves clean markdown from URLs when web_fetch fails, converting pages via Jina AI reader service with automatic retry, useful for 403, blocked, paywall, timeout, JavaScript-rendering errors, empty content, or when explicitly suggested.

Creates Skills for Claude. Use when users request creating or updating skills, need skill structure guidance, or mention extending Claude's capabilities through custom skills.

This skill performs zero-shot univariate time series forecasting using the Reverso foundation model (NumPy/Numba CPU-only inference), activating when users provide time series data and request forecasts, predictions, or extrapolations, and supporting Reverso Small (550K params).

Converts raster images (photos, paintings, illustrations, line art) into SVG vector reproductions. Use to vectorize, trace, or convert images to SVG, or to faithfully reproduce artwork as vector graphics.

Essential for all skill development, this tool automatically version controls every skill file modification for rollback or comparison. Use it after init_skill.sh, after every str_replace/create_file, and before packaging.

Control Spotify playback and manage playlists via MCP server. Use when user requests playing music, controlling Spotify, creating playlists, searching songs, or managing their Spotify library.

Executes programs on a compiled transformer stack machine where every instruction fetch and memory read uses a parabolic attention head. It demonstrates that transformer attention + FF layers can implement a working computer.

Exhaustive problem space exploration using the MIT Synthetic Neurobiology "tiling tree" method. It partitions a problem into MECE (Mutually Exclusive, Collectively Exhaustive) subsets recursively via parallel subagents, then evaluates leaf ideas against specified criteria.

Generates optimized instructions for Claude (Project instructions, Skills, or standalone prompts). Use when users request creating project setups, writing effective prompts, building Skills, or need guidance on instruction types for Claude.ai.

Enables GitHub repository operations (read/write/commit/PR) for Claude.ai chat environments. Use for GitHub commits, repository updates, DEVLOG persistence, or cross-session state management via GitHub branches. Not needed in Claude Code.

Generates WAFFLES Declarations for social media posts, providing preemptive lists of what a post does NOT say. Use it to prevent misinterpretation, clarify posts, or add disclaimers for controversial/nuanced takes, especially when WAFFLES is mentioned.

DEPRECATED - Use browsing-bluesky skill instead. This skill sampled and analyzed the Bluesky firehose to identify trending topics and content clusters, used for queries about network trends or real-time topic clusters.

Orchestrates parallel API instances, delegated sub-tasks, and multi-agent workflows with streaming and tool-enabled delegation patterns. Use for parallel analysis, multi-perspective reviews, or complex task decomposition.

Foveated vectorization for SVGs with 4-zone selective detail, akin to phone portrait mode but vectorized. It combines Claude vision annotations, MediaPipe segmentation/landmarks, and optional saliency.

Browse Bluesky content via API and firehose, including searching posts, fetching user activity, and analyzing accounts. It supports authenticated access for personalized feeds, ideal for research, monitoring, and trend analysis.

Maintains a structured task list for the current session, used proactively when a request requires 3+ distinct steps, multiple items, or complex work benefiting from explicit progress tracking. Storage persists via Muninn config across container death, adapted from Claude Code's TodoWrite tool.

In-process semantic search over text files or in-memory strings, using Gemini embeddings via the CF AI Gateway. It's ideal for fuzzy/conceptual searches where exact keyword matching would fail, complementing other search tools.

Browser automation via webctl CLI in Claude.ai containers with authenticated proxy support. Use when users mention webctl, browser automation, Playwright browsing, web scraping, or headless Chrome in container environments.

Universal environment variable loader for AI agent environments. Loads secrets and config from Claude.ai, Claude Code, OpenAI Codex, Jules, and standard .env files.

Discover and load skills on demand from /mnt/skills/user/. This tool helps you find capabilities, get full skill descriptions, or browse the catalog using 'list', 'search', and 'show' commands.

Write effective instructions for Claude, covering project instructions, standalone prompts, and skill content. This skill helps users improve communication with Claude by applying writing principles, model-aware calibration, and format selection.

This skill provides a file-upload bridge for Claude Code on the Web, which lacks native file mounting. It creates a throwaway GitHub branch for users to drop files via the web UI, fetching them locally for the session, ideal for uploading or sharing needed files.

Guidance for asking clarifying questions when user requests are ambiguous, have multiple valid approaches, or require critical decisions. Use when implementation choices exist that could significantly affect outcomes.

Closes a GitHub issue with a synthesis comment as a flowing graph, validating it, posting the comment, and then running a detached pluggable callback. Use this when closing an issue should capture learning, not just diffs, and post-close work shouldn't block the close acknowledgment.

Generates git patch files from codebase modifications for local application. Use when the user mentions patch, diff, export changes, bring changes back, apply locally, or after editing uploaded codebases.

Install and drive Google's Antigravity CLI (agy) as a non-interactive sub-agent. Use when orchestrating agy, running Antigravity agents from a script or sandbox, delegating a task to Google's agent harness, or wanting a Gemini-backed peer agent alongside Claude.

Generates a pre-change blast-radius report for a symbol or file, identifying affected sites by feature or package. It scans references and non-parsed files, useful before refactoring, renaming, or deleting in external repositories.

This tool provides image processing capabilities. Use it when users upload images for manipulation, format conversion, resizing, optimization, analysis, applying effects, or any image-related task, including working with screenshots and visual assets.

Access GitHub repositories in containerized environments using REST API and credential detection. Use when git clone fails, or for accessing private repositories/writing files via API.

Generates progressive disclosure indexes for GitHub repositories, to be used as Claude project knowledge. Use when setting up projects referencing external documentation, creating searchable indexes of technical blogs or knowledge bases, combining multiple repos into one index, or when the user mentions index, github repo, project knowledge, or documentation reference.

Cross-context adversarial review for deliverables before shipping. Use when producing blog posts, technical recommendations, analysis briefs, code, or any artifact where accuracy matters more than speed.

Composes single-file HTML artifacts (like reports, presentations, and flowcharts) from a JSON spec, replacing manual HTML/CSS/JS. Use to compare options side-by-side, request HTML versions of documents, or create flows.

This skill provides a first-encounter codebase orientation, chaining structural inventory and feature synthesis into an EDA workflow for unfamiliar repositories. Use it to quickly understand a new codebase, especially when asked to explore or understand a recently cloned repository.

Installs skills from github.com/oaustegard/claude-skills into /mnt/skills/user. Use when the user mentions install, load, add, update, refresh skills, or references a skill not currently installed.

Text-prompted image zone detection using TIPSv2 B/14 on CPU. It produces focus_targets / focus_edges bbox lists from natural-language labels, enabling automatic foreground/background separation without manual bbox annotation.

A DAG workflow runner that encodes control flow in code, not prose. It's designed for procedures with 3+ steps involving branching, retries, or enforced validation, using constructs like `when=`, `validate=`, and `retry_until=`, and supports parallel execution and checkpoint resume.

Specialized Preact development skill for standards-based web applications with native-first architecture and minimal dependency footprint. Use when building Preact projects, particularly those involving data visualization, interactive applications, single-page apps with HTM syntax, Web Components integration, CSV/JSON data parsing, WebGL shader visualizations, or zero-build solutions with vendored.