# agent-loop-skills: a set by gaasher

The gaasher/agent-loop-skills repository on GitHub: 26 cards with a description and an install command. Inside: 25 skills and 1 plugin. They install into Claude Code, Codex and Cursor.

26 cards

- HTML: https://affpapa.org/en/nevibecode/by/gaasher/agent-loop-skills
- Instructions for agents: https://affpapa.org/en/nevibecode/agents.md

- [literature-search](https://affpapa.org/en/nevibecode/skills/gaasher-literature-search.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — literature-search searches academic papers across Semantic Scholar and arXiv, walks citation trees, and extracts snippets using only the Python standard library.
- [data-analysis](https://affpapa.org/en/nevibecode/skills/gaasher-data-analysis.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — The data-analysis skill explores a dataset by testing hypotheses and writes up findings backed by reproducible numbers.
- [red-team](https://affpapa.org/en/nevibecode/skills/gaasher-red-team.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — red-team finds weaknesses in content filters, classifiers and APIs through adversarial stress-testing — generates attack inputs and builds a catalogue of failures.
- [prompt-optimize](https://affpapa.org/en/nevibecode/skills/prompt-optimize.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — prompt-optimize is a Claude Code skill that iteratively refines an existing prompt against your evaluation command, keeping only edits that improve the score.
- [anomaly-investigation](https://affpapa.org/en/nevibecode/skills/anomaly-investigation.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — anomaly-investigation is a Claude Code skill that investigates a known spike, drop, or outlier by testing possible causes against data.
- [power-analysis](https://affpapa.org/en/nevibecode/skills/gaasher-power-analysis.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — power-analysis is a Claude Code skill for sizing a two-group study and drafting its preregistration.
- [claim-verify](https://affpapa.org/en/nevibecode/skills/claim-verify.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — claim-verify is a Claude Code skill that checks data-backed claims in draft reports against raw datasets.
- [literature-survey](https://affpapa.org/en/nevibecode/skills/literature-survey.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — literature-survey is a Claude Code skill that maps research sources against claims.
- [optimize-loop](https://affpapa.org/en/nevibecode/skills/optimize-loop.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — optimize-loop is a skill that runs an edit-measure-revert loop for code modules or SQL queries, keeping changes only when correctness holds and a cost metric falls.
- [research-proposal](https://affpapa.org/en/nevibecode/skills/gaasher-research-proposal.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — Refines an existing research proposal through iterative literature checks, rubric-based scoring, and targeted revisions until it passes.
- [research-question](https://affpapa.org/en/nevibecode/skills/research-question.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — Turn a broad topic into a small set of research questions, then score and revise them.
- [karpathy](https://affpapa.org/en/nevibecode/skills/karpathy.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — karpathy — a Claude Code skill that turns the agent into an autonomous ML researcher: it edits training code, runs it, and keeps only the changes that lower a single scalar metric.
- [dueling-autoresearch](https://affpapa.org/en/nevibecode/skills/dueling-autoresearch.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — dueling-autoresearch pits two distinct technical approaches against each other on a shared metric.
- [plan-loop](https://affpapa.org/en/nevibecode/skills/plan-loop.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — plan-loop is a Claude Code skill that turns engineering requests into task-by-task plans before coding starts.
- [purple-team](https://affpapa.org/en/nevibecode/skills/purple-team.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — purple-team runs attack and defense as one loop: red-team finds failure classes in a filter or prompt you own, blue-team patches it, a fresh pass checks the patch.
- [hypothesis-gen](https://affpapa.org/en/nevibecode/skills/hypothesis-gen.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — hypothesis-gen generates testable research hypotheses and grounds them against published scientific literature using an agentic loop.
- [tournament-autoresearch](https://affpapa.org/en/nevibecode/skills/tournament-autoresearch.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — The tournament-autoresearch skill pits architecture proposals against each other before an ML experiment runs.
- [swe-loop](https://affpapa.org/en/nevibecode/skills/swe-loop.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — swe-loop is a Claude Code skill that breaks coding requests into tasks, has an Engineer implement them, and QA verify tests and quality.
- [scientific-writer](https://affpapa.org/en/nevibecode/skills/scientific-writer.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — scientific-writer refines scientific drafts through iterative critique and peer review.
- [alpha-evolve](https://affpapa.org/en/nevibecode/skills/alpha-evolve.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — The alpha-evolve skill explores many model-code variants in parallel, scores them through training runs, and keeps diverse strong candidates.
- [blue-team](https://affpapa.org/en/nevibecode/skills/blue-team.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — blue-team runs a fix loop over a concrete list of failures: one root-cause class per iteration, a revert on any regression, and a pull request at the end.
- [scientific-figure](https://affpapa.org/en/nevibecode/skills/scientific-figure.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — Builds publication-ready scientific figures from data or prompts through an adversarial generator-critic loop evaluated against a strict rubric.
- [tabular-cleanup](https://affpapa.org/en/nevibecode/skills/tabular-cleanup.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — The tabular-cleanup skill cleans messy CSV, TSV, Parquet, Excel, or JSON tables against an agreed data contract.
- [exploratory-autoresearch](https://affpapa.org/en/nevibecode/skills/exploratory-autoresearch.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — exploratory-autoresearch runs an autonomous ML experiment loop that samples wildly different designs first, then switches to picking one move per round.
- [ml-autoresearch](https://affpapa.org/en/nevibecode/skills/ml-autoresearch.md) — skill · gaasher · Claude Code, Codex, Cursor +4 · ★ 185 repo stars — ml-autoresearch drives an autonomous machine learning loop where model diagnostics and research papers shape each code iteration.
- [agent-loops](https://affpapa.org/en/nevibecode/plugins/agent-loops.md) — plugin · gaasher · Claude Code · ★ 185 repo stars — Run agent loops that propose a change, score the result, and keep or revert it.

## Install the whole set (Claude Code / Skills CLI)

```bash
/plugin marketplace add gaasher/Agent-Loop-Skills
```

## Narrow this list
Parameters: `?platform=<claude-code|codex|cursor|copilot|gemini-cli|windsurf|…>`, `&topic=<…>`, `&sort=installs|stars|new|updated` — values in https://affpapa.org/en/nevibecode/agents.md

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
