By cailmdaley
Drive autonomous iteration loops for research tasks, generate publication-quality data visualizations and scientific code, fetch arXiv full-text and BibTeX citations, create images via Gemini, and stream text-to-speech.
Speak text aloud using local TTS. Use when the user wants to hear something read out loud, spoken, voiced, narrated, or audibly rendered. Triggers on: "read this aloud", "say this", "speak", "out loud", "tell me [verbally]", "narrate", "voice this", "hear this", "read to me", "TTS", or any request to audibly render text.
Autonomous loop iteration toward a desired state. You are inside a ralph loop — your spec is in the system prompt. Survey, contribute, update state discoverably, exit. Activated automatically inside ralph loops. Triggers: "ralph-loops", "ralph", "ralph loop", "iterate", "autonomous loop".
Consult the other coding agent — Claude Code calls Codex, Codex calls Claude Code. Use for second opinions, code review, consensus, or dispatching tasks to the other agent. Two modes: one-shot for quick questions, tmux session for extended collaboration. Triggers: "confer", "ask claude", "ask codex", "second opinion", "get a review from", "use claude for", "use codex for", "dispatch to".
Draft a ralph constitution — a markdown spec describing a desired state for autonomous iteration. Study the problem space, shape the spec interactively, then launch the ralph loop. Use for any work where adaptation matters more than a fixed plan: science, refactoring, exploration, creative work. Triggers: "constitution", "constitute", "ralph spec", "set up a ralph", "create a ralph", "write a spec".
Create data visualizations, design figures, and plot analysis results. Specialized for observational cosmology (spectra, covariances, posteriors, sky maps) but capable of creative/expressive output for outreach. Use when: plotting data, designing figures for papers/talks, visualizing uncertainty, creating publication-quality graphics, or any task involving matplotlib/seaborn/plotly. Triggers on: "plot", "figure", "visualize", "chart", "graph", "histogram", "scatter", "heatmap", "contour", "power spectrum", "corner plot", or any request to create or improve data graphics.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Agent skills for research work — autonomous iteration, scientific visualization, bibliography management, image generation, and more. Currently packaged as a Claude Code plugin, but the skills themselves are just markdown files that any coding agent can use.
Read each skill before you run it. These skills shape how your agent works. ralph-loops and confer run agents with no permission gates — they bypass all approval checks and sandboxing. Understand what a skill does before activating it.
These reflect one person's workflow, not universal best practices. Fork the repo, delete what doesn't fit, rewrite what almost fits. The best skills are the ones you've made your own.
# Claude Code
claude plugin marketplace add cailmdaley/skills
claude plugin install research-skills
# Codex (no plugin system — symlink the skills directory)
git clone https://github.com/cailmdaley/skills.git ~/.local/share/skills
ln -s ~/.local/share/skills/skills/* ~/.agents/skills/
Two skills that work together. constitution is the design document — a markdown spec describing the desired state, written to outlast any single iteration. ralph-loops is the execution engine — fresh-context loops that survey reality, contribute, and exit, repeating until the constitution is satisfied.
Use the constitution skill to draft a spec: "use the constitution skill to help me write a spec for [your goal]." Then launch with the included ralph script, which runs in a tmux session with support for both Claude Code and Codex backends. Attach to watch, steer, or update the spec mid-run.
Pattern adapted from Geoffrey Huntley's Ralph Wiggum.
Consult the other coding agent — Claude Code calls Codex, Codex calls Claude Code. One-shot mode for quick questions, reviews, or dispatched tasks; tmux session mode for extended back-and-forth collaboration.
A decision framework synthesizing Bertin, Cleveland, Tufte, Cairo, Wilke, and Knaflic. Routes from data type to visualization form, with cosmology-specific conventions (power spectra, covariance matrices, triangle plots, sky maps). Includes a visualization catalog, color palette reference, and design system.
Opinionated approach to research code: conceptual density, concise conditionals, natural line breaking. Write it right the first time — clean, correct, no post-hoc cleanup needed.
Read full paper source from arXiv, fetch BibTeX from NASA ADS, generate citation keys, append to your .bib file. Requires $ADS_API_TOKEN.
Generate and edit images using Gemini CLI with persistent visual memory via KV cache warmth. Two warm caches (Gemini's KV cache + Claude's context) enable iterative refinement. Includes transparency extraction via difference matting.
Requires Gemini CLI with the nanobanana extension.
Speak text aloud using pocket-tts server with streaming audio. macOS only.
Requires: pip install pocket-tts, brew install ffmpeg, a voice file at ~/.config/pocket-tts/default-voice.wav.
Licensed under the Apache License, Version 2.0. Reuse implies no affiliation with or endorsement by Cail Daley; listings on third-party directories are not authorized without permission.
npx claudepluginhub cailmdaley/skills --plugin research-skillsfelt — fiber-based thinking substrate for Claude Code. Sessions start with active fibers; the fiber library stays visible throughout.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review
Design fluency for frontend development. 1 skill with 23 commands (/impeccable polish, /impeccable audit, /impeccable critique, etc.) and curated anti-pattern detection.