research-skills

Use this skill when the user asks to "fan out agents", "spawn agents", "use an agent team", "parallel agents", "delegate to subagents", "act as team lead", "orchestrate implementation", "fan out teammates", or when a task decomposes into 2+ independent work items (parallel exploration, one implementer per issue, a multi-lens PR review panel). Covers the basic mechanics of spawning, briefing, supervising, and synthesizing subagents and teammates, plus the hard cap on fleet size.

Use this skill when the user asks to "explore this codebase", "understand this repo", "map the architecture", "review our stack", "get up to speed", "break ground" on an unfamiliar project or research field, "what is the state of this project", "audit the architecture", or before planning changes in a codebase you have not worked in. Produces a verified reconnaissance report before any design or edit.

Use this skill when the user asks to "implement this change", "fix this issue", "add this feature", "make this change properly", "ship this", or for any ordinary single-PR development task: one issue, one branch, one pull request. For multi-phase features use epic-dev; for unfamiliar code run codebase-onboarding first; for failing behavior use the debugging skill.

Execute the project epic development workflow for Codex. Use when the user asks for "epic dev", "epic-dev", "/epic-dev", "run the epic workflow", "create an epic", "resume the epic", "--resume", "--next-phase", "--finalize", phased feature delivery, sprint phases, GitHub sub-issues, or git worktrees for multi-phase implementation.

Use this skill when the user asks to "plan this feature", "write an implementation plan", "design before coding", "think through the approach", "review this plan", or before any multi-file change, architectural decision, experiment with a go/no-go outcome, or strategic pivot. Produces a verifiable plan with pre-registered decision gates and an explicit list of open judgment calls.

Install, update, or audit general user-level AI coding instructions across Claude Code, OpenAI Codex, GitHub Copilot CLI, and Cursor. Use when the user asks to set up global instructions, personal AGENTS.md or CLAUDE.md defaults, configure instructions for multiple agent systems, avoid repeated downstream project rules, install model-routing defaults, or migrate Claude-only user rules to cross-agent user scope.

This skill should be used when the user asks to "update rules", "sync rules", "update AGENTS.md", "update CLAUDE.md", "refresh rules", "update project config", "sync templates", "update vibe rules", "check for rule updates", "update user rules", "update global config", "refresh AGENTS.md", "refresh CLAUDE.md", "sync .rules directory", "what's new in templates", or wants to update existing agent/rules configuration from the latest templates without overwriting customizations.

This skill should be used when executing the epic-dev workflow, creating epic branches, managing sprint phases, working with git worktrees for phased feature development, or when the user mentions "epic dev", "epic-dev", "/epic-dev", "epic workflow", "sprint phases", "phased development", or "git worktree workflow".

Use this skill when the user asks to "debug", "find the root cause", "figure out why X fails", "investigate this bug", "this test is flaky", "the build broke", "production incident", "why is the output wrong", or when any observed behavior contradicts expected behavior. Covers the reproduce-isolate-prove-fix-verify loop, anti-shortcut gates, and numerical/scientific debugging.

This skill should be used when the user says "security audit", "check for vulnerabilities", "security review", "harden project", "dependency audit", "credential scan", "check for secrets", "scan for secrets", "OWASP review", "security checklist", "audit dependencies", "find vulnerabilities", or wants to review their project for security issues, exposed credentials, or vulnerable dependencies.

Use this skill for "initialize project", "set up project", "scaffold project", "create AGENTS.md", "create CLAUDE.md", "set up rules", "create context directory", "vibe rules", "project templates", "init new project", "set up development structure", "create .rules", "create .context", "create ADR", "architecture decision records", "set up GitHub labels", "default issue labels", "project scaffolding", "set up pre-commit hooks", or when the user wants to initialize a new project with cross-agent development templates, structured documentation, and an optional default GitHub label set.

Review pull requests and recent code changes across specialized lenses: general code quality, tests, error handling and silent failures, comments and docs, type design, and simplification. Use when the user asks to review a PR, review recent changes, check tests, check error handling, check comments, analyze types, simplify/refine an implementation, run "review-pr", or before merging phase PRs in the epic-dev workflow.

This skill should be used when the user asks to "review a grant", "review my proposal", "score this grant", "evaluate my specific aims", "critique my research strategy", "review as an NIH reviewer", "review as an NSF panelist", "give me reviewer feedback", "check my grant proposal", "review my R01", "review my K99", "evaluate my CAREER proposal", "run a mock study section", "review my resubmission", "review this PDF", "check my proposal PDF", "analyze my grant layout", or mentions grant review, proposal critique, NIH scoring, NSF panel review, study section feedback, or proposal PDF review.

This skill should be used when the user asks to "write a grant proposal", "draft specific aims", "write a research strategy", "create an NIH proposal", "create an NSF proposal", "write a significance section", "write an innovation section", "write an approach section", "draft a DP2 essay", "write an R01", "write an R21", "write a K99", "write an R03", "write a K08", "write a K23", "write an F31", "write an F32", "write a CAREER proposal", "write preliminary data", "write rigor and reproducibility", "draft potential problems and alternatives", "write a budget justification", "respond to reviewer comments", "write a resubmission introduction", "strengthen my specific aims", "format grant text", or mentions grant writing, proposal drafting, specific aims, research strategy sections, or any NIH/NSF mechanism.

This skill should be used when the user asks to "generate an AI figure", "make a full AI-generated figure", "make a poster figure", "make a presentation figure", "generate a figure with AI", "create a substrate image", "AI background for a figure", "create a poster background", or wants a pictorial substrate (brain scene, microscope setup, anatomical scene, lab apparatus) generated by AI and then composited with labels and arrows in code. Defaults to the Codex CLI image_gen tool (preferred when `codex login` is configured) or the OpenAI Images API. Hard ceiling is any figure that needs data plots, axis numerals, equations, multi-arrow flowcharts, or more than 5 labeled elements should use scientific-figure or svg-figure instead.

Use when the user asks to build a programmatic SVG schematic in Python: flowcharts, boxes and arrows, auto-fit SVG text, mm-precise diagrams, edge-snapped arrows, tangent-correct curve arrowheads, orthogonal or Manhattan routing, deterministic z-order layers, labeled groups, bracket groupings, leader-line annotations, or strict Canvas.save validation for text overflow. Use figure-qa instead for after-the-fact validation of arbitrary hand-authored or exported SVGs. Built on drawsvg, svgpathtools, and fontTools; output SVGs can be composed by scientific-figure.

Use this skill to QA a scientific figure for journal compliance, alignment, palette correctness, and label legibility. Triggers on "QA this figure", "check this figure", "review my figure", "is this figure paper-ready", "validate figure", or when invoked proactively after a figure-generation skill produces output (unless the caller passes no-qa). Dispatches on input type (SVG, raster PNG/JPG/TIFF, Python plot script, or composed-figure directory) and runs the right programmatic checks plus a VLM rubric judgment pass.

Use this skill to review grant proposal figures for compliance, resolution, accessibility, and quality. Triggers on "check grant figures", "review proposal figures", "figure QA for grant", "NIH figure requirements", or when preparing a grant for submission.

Use this skill for "review this paper", "review this manuscript", "peer review", "review my paper", "critique this manuscript", "review this submission", "give me feedback on my paper", "check my methods", "review my statistics", "review as a peer reviewer", "evaluate this manuscript", "review this PDF", or mentions manuscript review, peer review, paper critique, or methodological review.

This skill should be used when the user asks to "create a presentation", "make slides", "build a slide deck", "create a talk", "make a keynote", "create a Reveal.js presentation", "generate presentation slides", "make a conference talk", "create a lecture", "build a poster presentation", "create presentation JSON", or mentions presentations, slides, slide decks, Reveal.js, talk preparation, conference presentations, or lecture slides.

This skill should be used when the user asks to "create a figure", "make a scientific figure", "create a paper figure", "make a figure for my paper", "make a figure for my manuscript", "compose figure panels", "assemble figure panels", "combine figure panels", "make a multi-panel figure", "figure composition", "export a figure", "generate a PDF figure", "create a Nature-style figure", "make a journal figure", "make a publication figure", or "create a figure for my grant". The skill composes multi-panel figures at exact journal dimensions (Nature 89/183 mm, Science 55/120 mm, Cell 85/174 mm, PNAS 87/180 mm) using svgutils, validates font sizes against journal minima before export, and writes PDF/PNG via Inkscape when available or cairosvg otherwise.

This skill should be used when the user asks to "create an SVG figure", "make a schematic", "draw a diagram", "create a schematic diagram", "draw a flowchart", "draw a process flow", "draw a workflow", "draw a workflow diagram", "make an SVG schematic", "create a process diagram", "create a pipeline diagram", "create a block diagram", "draw a system diagram", "system architecture diagram", or wants a hand-authored or programmatic SVG with shapes, arrows, and labels that the figure-qa agent can verify. **For new Python-driven figures, route to `[[svg-primitives]]` instead** — this skill is the conventions and hand-authoring reference. Outputs are SVG files that can be loaded as panel sources by the scientific-figure composer.

This skill should be used when the user asks to "make an icon", "generate an icon", "create a scientific icon", "make a transparent icon", "make a minimal icon", "icon for a figure", "icon for a paper", "generate an icon set", "batch icons", "icon batch", "batch generate icons", or wants flat scientific icons (brain, neuron, DNA, EEG cap, etc.) in the Nature/Science journal style with transparent backgrounds. Generates PNG icons via the Codex CLI image_gen tool (preferred) or the OpenAI Images API (fallback), then applies transparency via a Pillow threshold by default or rembg + BiRefNet (opt-in, higher edge quality).

This skill should be used when the user asks to "plot data", "make a plot for a paper", "make a publication plot", "create a journal-quality plot", "matplotlib for a figure", "use seaborn", "use plotnine", "use ggplot in Python", "improve plot quality", "style my plot", "clean up my figure", "my matplotlib plot looks ugly", "apply journal style to existing script", "fix matplotlib defaults", "use SciencePlots", "make a Nature-style plot", "make an IEEE-style plot", "make a paper-ready chart", "which plotting library should I use", or wants guidance on choosing between matplotlib / seaborn / plotnine / plotly / pyvista and applying journal-quality defaults. Produces SVG-output Python scripts whose panels are consumed by the scientific-figure composer.

Use this skill for "humanize this text", "remove AI tells", "make this less AI-sounding", "strip AI patterns", "de-AI my writing", "reduce AI-isms", "final natural-writing pass", "check for AI vocabulary", "em dash overuse", "rule of three overuse", "AI vocabulary", "signs of AI writing", or when polishing prose (papers, grants, lit reviews, abstracts, cover letters, responses to reviewers) to remove signs of AI-generated text while preserving meaning and discipline-appropriate conventions.

Use this skill for "literature review workflow", "multi-phase lit review", "direction paper", "review paper protocol", "strand-based literature review", "citation-grounded review", "systematic lit review with paper cards", "build a lit review corpus", "lit review pipeline", "orchestrate a literature review", "research directions document", "write a literature review", "synthesize papers", "thematic review", "narrative review", "systematic review", "scoping review", "gap analysis", or when the user wants either a rigorous multi-phase citation-traceable lit review or a single-pass thematic synthesis for an Introduction/Background section.

Use this skill for "format manuscript", "prepare for submission", "journal formatting", "LaTeX template", "submission checklist", "format references", "BibTeX", "author guidelines", "page limits", "format for Nature", "format for IEEE", or when the user wants to format a manuscript for journal submission.

Use this skill for "write a paper", "draft manuscript", "write introduction", "write methods section", "write results", "write discussion", "write abstract", "structure a paper", "academic writing", "write for journal", or when the user wants to draft or revise sections of an academic manuscript.

Use this skill for "convert to BIDS", "BIDS format", "create BIDS dataset", "BIDS sidecar", "participants.tsv", "events.tsv", "EEG BIDS", "EMG BIDS", "MEG BIDS", "fMRI BIDS", "BIDS validator", "channel locations", "electrode positions", "BIDS metadata", or when the user wants to convert neuroscience data to Brain Imaging Data Structure (BIDS) format or validate BIDS compliance.

Use this skill for "design experiment", "create PsychoPy experiment", "stimulus presentation", "experiment protocol", "timing validation", "trial structure", "block design", "event-related design", "PsychoPy builder", "create stimuli", "LSL markers", "Lab Streaming Layer", "event markers", "trigger codes", or when the user wants to design or implement a neuroscience experiment.

This skill should be used when the user says "set up CI", "create GitHub Actions", "scaffold CI pipeline", "add CI/CD", "configure continuous integration", "create test workflow", "create release workflow", "add pre-commit hooks", "set up linting pipeline", "configure ruff in CI", "configure biome in CI", "add typo checking", or wants to add, update, or customize CI/CD pipelines for their project. For initial project setup including basic CI, see init-project. Use ci-scaffolding for adding or customizing CI/CD in existing projects.

This skill should be used when the user says "create Dockerfile", "dockerize project", "Docker packaging", "container setup", "multi-stage build", "Docker Compose", "containerize application", "create docker-compose.yml", "create .dockerignore", "optimize Docker image", or wants to containerize their project or create Docker configurations.

This skill should be used when the user says "process documents", "extract text from PDF", "OCR this document", "convert PDF to markdown", "extract emails from documents", "parse document", "document conversion", "batch OCR", "extract structured data from PDF", "read PDF", "extract tables from PDF", "convert Word document", "convert docx to markdown", or wants to extract, convert, or process documents and scanned images.

Use this skill for "search for papers", "find citations", "look up a DOI", "get BibTeX", "download PDF", "convert PDF to markdown", "find canonical papers", "convert identifiers", "batch download papers", "configure opencite", "literature review", "find related papers", "what papers cite this", "export references", "read this paper", or mentions of opencite, Semantic Scholar, OpenAlex, PubMed, academic literature search, citation management, or paper retrieval.