peteromallet/Arnold

General-purpose planning and execution harness for LLMs — structured phases, critique, gating, and review

This repository also configures its own agents. See what Arnold tells them →

98Stars on the repository
36Mods indexed here, across every type
2d agoLast push, which is what freshness is scored on
customA LICENSE file GitHub cannot name, so bodies are not copied

peteromallet/Arnold

Skill Claude CodeCodex

The merged babysitting protocol for the fix-the-fixer loop: a once-per-hour check-in that (1) checks for failures and repeated fixer misses, (2) runs the fixing loop — DeepSeek Flash swarm to understand, Grok for the recommendation, DeepSeek to implement, all inside sub-agents, biased toward fixing the fixer — and (3)…

not rated 98 +1 2d ago A 232 tokens

subagent-launcher

27

peteromallet/Arnold

Skill Claude CodeCodex

Launch an external model as a subagent for a second opinion, adversarial review, or delegated work. Default pathway is an agentic DeepSeek / Kimi / Zhipu GLM omp subagent (file/web/terminal tools, one process or fanned out N-wide); also Codex (GPT-5.5), Claude via the Agent tool, and Grok via omp (same x.ai token as…

not rated 98 +1 2d ago A 121 tokens

deliberation

29

peteromallet/Arnold

Skill Claude CodeCodex

The deliberation pipeline refines an input idea through layered critique: question generation, a human answer gate, and three abstraction layers (high strategic → mid architectural → low ground-truth), each running a fan-out panel of critics whose outputs are skeptically synthesized into a revised plan with a…

not rated 98 +1 2d ago A 0 tokens

folder-audit

30

peteromallet/Arnold

Skill Claude CodeCodex

Walk a directory tree and audit whether each folder's children belong at the right level of abstraction. Produces a structured JSON audit plus a nested Markdown annotated tree.

not rated 98 +1 2d ago A 38 tokens

simplify-writing

31

peteromallet/Arnold

Skill Claude CodeCodex

Fan out DeepSeek Pro critics and a Kimi editor to iteratively simplify a piece of writing.

not rated 98 +1 2d ago A 24 tokens

vibecomfy-executor

32

peteromallet/Arnold

Skill Claude CodeCodex

Arnold pipeline that classifies a user query, optionally researches via Hivemind/Banodoco, optionally edits a ComfyUI workflow, and always emits a final reply.

not rated 98 +1 2d ago A 44 tokens

hello

33

peteromallet/Arnold

Skill Claude CodeCodex

A minimal Python-shaped Arnold workflow that greets and responds.

not rated 98 +1 2d ago A 0 tokens

At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: