Agent Loop Skills — Data skill for Claude Code
Loop until it's better — drop-in agentic loops (autoresearch, scientific writing, data analysis, code/SQL/prompt optimization, red-teaming) as open-standard Agent Skills.
How to install Agent Loop Skills
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open gaasher/Agent-Loop-Skills and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Agent Loop Skills does
Loop until it's better — drop-in agentic loops (autoresearch, scientific writing, data analysis, code/SQL/prompt optimization, red-teaming) as open-standard Agent Skills. Verification-gated; native on Claude Code, portable across Codex, Cursor & other Skills hosts.
Alternatives in Data
- LLM Redteam — LLM red-team corpus runner — fires categorized prompt-injection / jailbreak / system-prompt-leak / data-exfil 4.5k ★
- Autonovel — An autonomous novel writing pipeline, by Hermes Agent 1.5k ★
- Agentlas Storm — Run a force-robust Stormbreaker loop — route to real agents, execute a verified pipeline to completion 1.1k ★
README
agent-loop-skills
Loop until it's better — drop-in agentic **loops**, packaged as open-standard **Agent Skills**.
Autoresearch · scientific writing · data analysis · code/SQL/prompt optimization · red/blue/purple-teaming — each a *generic, reusable loop* you bind to **your own task at invocation time**, that iterates against a **real signal** until the work is actually better.
[](https://agentskills.io/specification) [](#compatibility)  [](CONTRIBUTING.md) [](LICENSE) [](https://github.com/gaasher/agent-loop-skills)
A real run. The tournament-autoresearch loop on a CIFAR-10 model under a fixed 5-epoch budget —
competing agents propose a change each step, a self-calibrating judge keeps the winners (green) and discards the regressions (gray):
0.734 → 0.798 val_acc, hands-off, 7 of 11 kept. Full ledger: showcase/tournament-autoresearch.
Far from SOTA by design — a deliberately tiny CNN at 5 epochs on a laptop GPU (Apple MPS). The demo is the loop's decision-making, not the absolute accuracy.
Why loops-as-skills
Two ideas collided in late 2025, and this repo lives in the overlap:
- **Skills became the
Related Skills
Autoresearch Skill Andrej Karpathy
Claude Code skill for autonomous, goal-directed iteration. /autoresearch builds a real-data benchmark harness,
Claude TDD Kit
Claude Code plugin marketplace: the developer day-loop plus a gated TDD build pipeline. Two independent plugin
Multi Agent Evolution Framework
Multi-Agent Coding Orchestration Framework, Governed, CLI-agnostic, with HarnessX inspired gated in-process to
Micro Skills
Stop writing 200-line skills that Claude skims. Convert any skill into a gated micro-pipeline with hard YES/NO
Activitywatch Analysis Skill
Claude Code skill for weekly productivity analysis using ActivityWatch data. Calculates focus scores, detects
Database Sqlite
Skill: Database (SQLite via better-sqlite3)
Related Agents
Burn It All Down
Adversarial anarchist reviewer. Use to pressure-test a plan, design, or architecture by turning fixed constrai
Cicd Redteam
Delegates to this agent when the user wants to integrate red teaming into CI/CD pipelines, set up continuous a
Maker Reviewer
MAKER Reviewer - Final quality gate that reviews completed work for completeness, gaps, and doc compliance. MU
