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-reflection-self-improvement/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-reflection-self-improvement)<a href="https://agentmods.dev/skills/marin-community/marin/scrub-reflection-self-improvement"><img src="https://agentmods.dev/badge/skills/marin-community/marin/scrub-reflection-self-improvement/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-reflection-self-improvement"><img src="https://agentmods.dev/badge/skills/marin-community/marin/scrub-reflection-self-improvement.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Agent Snooping · line 51 Skill enumerates or reads other installed skills. Access to other skills' SKILL.md files or the skills directory reveals prompt instructions, capabilities, and secrets that should be invisible to peer skills.Fix: Remove all code or instructions that list or read other skills' files or directories. Skills should operate independently; cross-skill access is a privilege escalation.
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 | $0.00027 | $0.00747 |
| Opus 5 | $0.00014 | $0.00374 |
| Sonnet 5 | $0.00005 | $0.00149 |
| Haiku 4.5 | $0.00003 | $0.00075 |
Grade A, and why
scrub-reflection-self-improvement 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 10d 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 — 61 lines — stays where its author put it; the contents beside it link to each section on GitHub.
scrub-reflection-self-improvement
Use this skill on scheduled scrub turns to identify and land high-leverage improvements in marin-community/marin.
Focus
- Look for improvements from recent issues, PR feedback, and recurring operational friction.
- Prefer one concrete implementation per run when feasible.
- If implementation is blocked, produce a concrete plan and capture follow-up work in GitHub.
Candidate Signals
- Repeated confusion in docs, recipes, or contributor workflows.
- Recurring failures or avoidable manual steps in experiments, scripts, and infra operations.
- Capability gaps that reduce the value of agent-assisted contributions.
Triage Checklist
Run a lightweight, repeatable scan before choosing work:
- Review recent open issues and open PRs for repeated friction clusters.
- Explicitly check for already-open scrub-generated issues/PRs touching the same area; prefer advancing or deferring to that existing artifact instead of creating a parallel one.
- Review the latest commits on
mainfor changes that imply follow-on docs/workflow updates. - Search
AGENTS.md,.agents/skills/, and docs for stale workflow guidance related to those clusters. - De-duplicate against existing issues/PRs before creating new artifacts.
When possible, prefer improvements that remove recurring operator time (for example, turning ad-hoc scrub judgment into explicit repeatable guidance).
Decision Heuristics
- Pick the highest-leverage change with the lowest coordination overhead.
- Treat open scrub-generated issues/PRs as first-class prior art during triage; if one already covers the candidate improvement, avoid opening a second artifact unless the new scope is clearly distinct.
- De-duplicate against existing issues/PRs before opening new work.
- When an improvement changes recurring workflow guidance, codify it in durable repo instructions:
AGENTS.mdfor cross-cutting agent behavior, or.agents/skills/for repeatable task workflows. - If no justified improvement exists now, choose a no-op outcome.
- Prefer direct implementation over opening new issues when the change is fully in-repo and low-risk.
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.
- 10d ago First seen · 61 lines · 27 tokens per session scan A 88a1f88c3b3c
scrub-reflection-self-improvement is a skill published in the GitHub repository marin-community/marin (3,548 stars, last pushed today), licensed Apache-2.0. It adds 27 tokens to every session and 747 once invoked, about $0.0001 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…