Marin is an open-source research program, software platform, and community for developing foundation models such as large language models. Researchers use it for data preparation, tokenization, pretraining, posttraining, evaluation, and related experiments, including work on audio-text, DNA, and protein models. The catalogue entries are add-ons that support workflows around Marin.
Borrowing it
Nothing to install: this file belongs to marin-community/marin. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/marin-community/marin/main/.agents/skills/scrub-experiment-issue-tldrs/SKILL.mdgit clone --depth 1 https://github.com/marin-community/marinWrote 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/marin-community/marin/scrub-experiment-issue-tldrs)<a href="https://agentmods.dev/skills/marin-community/marin/scrub-experiment-issue-tldrs"><img src="https://agentmods.dev/badge/skills/marin-community/marin/scrub-experiment-issue-tldrs/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/marin-community/marin/scrub-experiment-issue-tldrs"><img src="https://agentmods.dev/badge/skills/marin-community/marin/scrub-experiment-issue-tldrs.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00031 | $0.00682 |
| Opus 5 | $0.00015 | $0.00341 |
| Sonnet 5 | $0.00006 | $0.00136 |
| Haiku 4.5 | $0.00003 | $0.00068 |
Grade A, and why
scrub-experiment-issue-tldrs 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 11d 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 — 64 lines — stays where its author put it; the contents beside it link to each section on GitHub.
scrub-experiment-issue-tldrs
Use this skill on scheduled scrub turns that maintain experiment issue summaries in marin-community/marin.
The Python selector script only picks which issues to inspect and provides thread context. All summary judgment, writing, and GitHub issue editing lives in this workflow.
Focus
- Keep experiment issues understandable to a technically strong newcomer who does not know the local project history.
- Prefer issues whose current body lacks a managed TL;DR block or whose existing summary is weak, stale, vague, or unlabeled.
- Treat closed issues as fully eligible. They often have the clearest conclusions and are good summary targets.
Managed Block Format
For each candidate issue, update or add exactly one managed issue-body block bounded by <!-- experiment-tldr:start --> and <!-- experiment-tldr:end -->.
Write the block as normal Markdown in this shape:
<!-- experiment-tldr:start -->
## Summary
One short newcomer-friendly summary paragraph.
### Helpful links
- <smallest useful set of links>
<!-- experiment-tldr:end -->
Writing Guidance
- Explain the setup, the investigation, and why it mattered.
- State the current conclusion, recommendation, or unresolved blocker in concrete language.
- Improve existing managed summaries whenever they are inaccurate, stale, vague, or miss the real conclusion.
- Improve unmanaged summaries too when the issue still lacks the
tldrlabel and the current body is not adequate. - Treat
250words as a soft cap for the summary section, not a target.
Helpful Links Guidance
- Keep the list short.
- Prefer decisive comments, W&B reports, follow-up PRs, linked issues, and similar artifacts that let a reader verify the summary quickly.
- Omit redundant or low-value links.
Label And Edit Guidance
- The selector script output is the source of truth for candidate order and provided thread context.
- Use
gh issue view --json <fields>,gh pr view --json <fields>, explicit narrow flags such as--comments, andgh apito inspect related issues, PRs, or comments when the provided context is not enough. - Skip issues whose body already matches the desired managed block content.
- The
tldrlabel means the issue now has an adequate newcomer-friendly summary plus enough supporting links to dig deeper. - Add the
tldrlabel when the issue now meets that bar. Remove it when the issue no longer meets that bar. - After updating an issue body, add a short
@dlwhcomment describing what changed.
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.
- 11d ago First seen · 64 lines · 31 tokens per session scan A e21eb9afbb9e
scrub-experiment-issue-tldrs is a skill published in the GitHub repository marin-community/marin (3,593 stars, last pushed today), licensed Apache-2.0. It adds 31 tokens to every session and 682 once invoked, about $0.0002 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…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
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…