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/refresh-fork/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/refresh-fork)<a href="https://agentmods.dev/skills/marin-community/marin/refresh-fork"><img src="https://agentmods.dev/badge/skills/marin-community/marin/refresh-fork/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/refresh-fork"><img src="https://agentmods.dev/badge/skills/marin-community/marin/refresh-fork.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, 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 52 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.
- medium Agent Snooping · line 278 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.
- medium Agent Snooping · line 273 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 | $0.03792 |
| Opus 5.5 | $0.00011 | $0.01517 |
| Sonnet 5.5 | $0.00005 | $0.00758 |
| Haiku 4.5 | $0.00003 | $0.00379 |
Grade A, and why
refresh-fork 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 14d 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 — 276 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Refresh a fork
Read AGENTS.md and the target descriptor. Select a source and replay the
fork's patches onto main-next when a rebase is needed. A fork_main selector
uses an existing stable source SHA.
For vllm-gpu, follow the GPU guide:
review and land source on main, build and test the wheels, then test and open
a Marin adoption PR. Other rebased forks test their staged sources before an
admin moves main. The refresh agent never force-moves a protected branch.
Refresh a group as one unit and one PR. Per-fork guidance lives beside this
file: docs/vllm.md covers the vllm/tpu-inference group and the GPU release
pipeline, and docs/xla.md covers the XLA PJRT fork, which is pinned outside
migration.toml entirely.
In local mode, ask before pushing fork branches, opening the PR, or filing an issue. The required end-to-end test needs no extra confirmation.
Read the Descriptor
Read the target fork's section in config/external/migration.toml. It gives:
upstream— the repo we rebase onto. Every fork has one.group— if present, refresh every section in the group together in one PR (read them all now); if absent, this pin refreshes alone.base_select— how to choose the new upstream base or existing fork source.pin— where the resolved pin is recorded (isolated_projectuv.lock,descriptor:<path>#<section>SHA, orrelease:<path>prebuilt wheel); drives the re-pin step.branch— the maintained fork branch this pin tracks. All current forks usemain. Rebase refreshes stage onmain-next;fork_mainselectors do not.e2e— the Marin end-to-end that validates the refresh.blocker_assignee— who owns the "can't migrate" issue.nuances— constraints a human must respect (torch pins, known-good ceilings).
The descriptor records how to migrate; the pin source it names holds the actual revision.
Outcome
- If no newer base is selected and no pin metadata needs repair, exit successfully with a no-op summary.
- On success for a descriptor or isolated-project rebase, create the rollback
and date tags described in
docs/promotion-protocol.mdfor each rebased fork, then open exactly one draft PR inmarin-community/marinfor the fork or group after the e2e passes. A grouped refresh re-pins every group section at its staged tip in that single PR. State the exactmain-nexttomainadmin promotion still required, request the descriptor'sblocker_assigneeas reviewer, and monitor the PR per.agents/skills/manage-pr/SKILL.md. - For
vllm-gpu, usedocs/vllm.mdfor the source-review handoff and the later Marin adoption PR. - On an unresolved blocker, do not open a PR. Create or update one
marin-community/marinissue assigned toblocker_assignee, titledFork refresh blocked: <fork> — <short reason>, with current pins, the selected base, branch names/SHAs if created, attempted fixes, the remaining failure, and artifacts.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 14d ago Changed · +1 lines 269c7daffcd6
- 15d ago Changed a5021e770c2e
- 17d ago Changed · -26 lines c35695f4289e
- 25d ago Changed · +9 lines b8471e61cfb7
- 1mo ago Changed · +7 lines 6bff64bfe3d9
- 2mo ago Changed · -50 lines 2404546fefe5
- 2mo ago First seen · 335 lines · 27 tokens per session scan A 650f9f6996b4
refresh-fork is a skill published in the GitHub repository marin-community/marin (4,074 stars, last pushed today), licensed Apache-2.0. It adds 27 tokens to every session and 3,792 once invoked, about $0.0001 per session on Opus 5.5. 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.
refresh-fork compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| refresh-forkmarin-community/marin | Skill | 4.1k | today | Apache-2.0 | A | 27 | Claude Code, Codex |
| merge-batchdiegosouzapw/OmniRoute | Skill | 83k | 5d ago | MIT | A | 46 | Claude Code |
| prowler-commitprowler-cloud/prowler | Skill | 15k | yesterday | Apache-2.0 | A | 44 | Claude Code, Codex |
| gh-auth-isolationgithub/gh-aw | Skill | 5.6k | today | MIT | A | 20 | Claude Code, Codex |
| github-skillzeenie-ai/OpenCompany | Skill | 1.1k | yesterday | MIT | A | 51 | Claude Code |
| comet-githubrpamis/comet | Skill | 3.3k | today | MIT | A | 72 | Claude Code, Codex |
| re0-mergeLilMGenius/paperthin | Skill | 1.4k | 4d ago | MIT | A | 70 | Claude Code |