Molt is a PyTorch-based reinforcement-learning framework for training software agents, including multimodal and multi-turn agents. It is intended for research using Ray for orchestration, vLLM for model rollouts, and NVIDIA AutoModel with FSDP2 for training.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/nvidia-nemo/labs-molt/simplicity-firstnpx skills add NVIDIA-NeMo/labs-molt --skill simplicity-firstgit clone --depth 1 https://github.com/NVIDIA-NeMo/labs-moltWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/nvidia-nemo/labs-molt/simplicity-first)<a href="https://agentmods.dev/skills/nvidia-nemo/labs-molt/simplicity-first"><img src="https://agentmods.dev/badge/skills/nvidia-nemo/labs-molt/simplicity-first.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00051 | $0.00629 |
| Opus 5 | $0.00026 | $0.00315 |
| Sonnet 5 | $0.00010 | $0.00126 |
| Haiku 4.5 | $0.00005 | $0.00063 |
Grade A, and why
simplicity-first scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Simplicity is the first principle
Human readability comes first; coding-agent traceability is the minimum gate. A human should understand the code in one pass. An agent must at least be able to trace a feature from CLI flag to executed branch, tensor/record, metric, and test without reconstructing hidden control flow. Every rule below is an instance of that ordering. These are hard correctness rules, not style preferences. A violation is a bug and must be fixed before the change ships.
-
Code a human can't follow at a glance is a bug. If a reviewer can't read a function top-to-bottom in one pass, restructure or delete it. Nesting, indirection, and clever constructs count against correctness — cleverness that costs comprehension is a defect, whatever it saves.
-
Too much / redundant code is a bug. Solve the problem in the fewest lines that stay readable. Prefer deleting code over adding it. A fix that adds more than ~20 lines for a problem statable in one sentence is suspect — find the smaller fix first.
-
Simplicity is the core engineering metric. When two designs both work, ship the one with less code, fewer concepts, fewer files. Never add config, record types, or return-shape changes "for the future".
-
No over-encapsulation. No new class / dataclass / helper / module for a single call site. A helper needs 3+ real call sites AND nontrivial logic — otherwise inline it. Never wrap trivial code. Never change a function signature or return shape to thread data that only one caller needs.
-
Simplicity is not deletion of capability. Features, performance knobs, and observability are intentional — do not remove them in the name of simplicity. Knobs default ON stay ON. Simplify the implementation, keep the behavior surface.
Checklist before finishing any change
- Would a human reading this cold understand it in one pass? That is the gate.
- Could this diff be half the size? If unsure, make it smaller.
- Any new class or file? Justify each with 3+ call sites, or delete it.
- Any signature / return-shape change? Verify every caller genuinely needs it.
- Comments: concise "why" only, 2-4 lines max, written for an external reader — no job ids, commit hashes, single-run metrics, or internal paths; keep upstream issue/PR links.
- One problem = one minimal diff. Do not batch unrelated "improvements".
- A "bug" that cannot trigger under the real recipes is not worth fixing.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago First seen · 50 lines · 51 tokens per session scan A bf4cd934475f
simplicity-first is a skill published in the GitHub repository NVIDIA-NeMo/labs-molt (1,017 stars, last pushed today), licensed Apache-2.0. It adds 51 tokens to every session and 629 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…