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/add-grant/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/add-grant)<a href="https://agentmods.dev/skills/marin-community/marin/add-grant"><img src="https://agentmods.dev/badge/skills/marin-community/marin/add-grant/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/add-grant"><img src="https://agentmods.dev/badge/skills/marin-community/marin/add-grant.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 | $0.00027 | $0.01605 |
| Opus 5 | $0.00014 | $0.00803 |
| Sonnet 5 | $0.00005 | $0.00321 |
| Haiku 4.5 | $0.00003 | $0.00161 |
Grade A, and why
add-grant 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 9d 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 — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: Add a user grant
Turn an access request into a reviewable Pulumi change. Every human principal is
KMS-encrypted, including IAP viewers on Cloud Run services. The change is never
applied here — a second person runs the
review-grant skill, merges, and runs pulumi up.
Read first:
infra/pulumi/README.md— the marin-iac stacks, the KMS key, and thepulumi upprerequisites.infra/pulumi/src/iac/gcp/iam_data.yamlheader — why humanuser:principals are encrypted and this file is public.
Grant surfaces
Decide which one the request needs before editing anything. A single request can touch both.
-
Shared project / resource GCP IAM — a role on the
hai-gcp-modelsproject, the KMS key, a Secret Manager secret, a GCS bucket, an Artifact Registry repo, or a service account (who may impersonate it). Lives ininfra/pulumi/src/iac/gcp/iam_data.yaml, applied by themarinstack ininfra/pulumi. Each humanuser:<email>principal is KMS-encrypted once in theprincipalsregistry; grants reference its opaquehuman-NNNID. Service accounts, groups, and domains stay plain strings. -
Deploy-target IAM — runtime, secret, repository, KMS, and IAP grants for Echo, EvalDash, Grafana, or Loom. Lives in that target's Python module under
infra/pulumi/src/iac/gcp/and is composed into themarinstack. Human grants reference the encrypted principal registry by opaque ID.
If you are unsure which surface a request means (e.g. "give Alice access to eval
results" could be an IAP viewer on evaldash, a roles/storage.objectViewer
grant on the record bucket, or both), ask before editing.
Collect the request
You need, per grant:
- Principal — an email for a person, or a
serviceAccount:/group:/domain:member for automation. Only personal emails get encrypted. - What they need access to — the specific resource, stated as a capability ("read the eval record bucket", "impersonate the ray autoscaler SA") rather than a raw role when the requester does not know GCP roles.
- Why / for how long — a one-line justification. If the access is temporary,
note it;
GcpIamConditioncan scope a grant with a CEL expiry, but prefer a follow-up removal PR unless the requester asks for an expiry.
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.
- 9d ago First seen · 141 lines · 27 tokens per session scan A 05cbb501aff6
add-grant is a skill published in the GitHub repository marin-community/marin (3,512 stars, last pushed yesterday), licensed Apache-2.0. It adds 27 tokens to every session and 1,605 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
gke-compute-classes
Configures, optimizes, and troubleshoots GKE ComputeClasses. Use when configuring Spot VMs with on-demand fallback, targeting specific accelerators (GPUs/TPUs) or machine families, restricting ComputeClass access, or debugging pending pods related to node pool auto-creation. Do not use for cluster-level Node Auto…
gke-workload-security
Audits, configures, and hardens workload-level security controls for Google Kubernetes Engine (GKE) applications and namespaces. Covers running cluster security audits (auditcluster.sh), configuring Workload Identity Federation (impersonation, KSA/GSA binding, and pod setup), enforcing Network Policies (default-deny…
gke-reliability
Improves GKE workload reliability, using PDBs, health probes, and topology spread constraints. Use when configuring GKE workload reliability, setting up PDBs, or configuring GKE health probes (liveness, readiness, startup). Don't use for disaster recovery setup or full cluster backups (use gke-backup-dr instead).
nemo-automodel-launcher-config
Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.
azure-mgmt-botservice-dotnet
Azure Resource Manager SDK for Bot Service in .NET. Management plane operations for creating and managing Azure Bot resources, channels (Teams, DirectLine, Slack), and connection settings. Triggers: "Bot Service", "BotResource", "Azure Bot", "DirectLine channel", "Teams channel", "bot management .NET", "create bot".
cloud-architect
Designs cloud architectures, creates migration plans, generates cost optimization recommendations, and produces disaster recovery strategies across AWS, Azure, and GCP. Use when designing cloud architectures, planning migrations, or optimizing multi-cloud deployments. Invoke for Well-Architected Framework, cost…