Borrowing it
Nothing to install: this file belongs to cloudrift-ai/emmy. 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/cloudrift-ai/emmy/main/.agents/skills/start-remote-server/SKILL.mdgit clone --depth 1 https://github.com/cloudrift-ai/emmyWrote 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/cloudrift-ai/emmy/start-remote-server)<a href="https://agentmods.dev/skills/cloudrift-ai/emmy/start-remote-server"><img src="https://agentmods.dev/badge/skills/cloudrift-ai/emmy/start-remote-server/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/cloudrift-ai/emmy/start-remote-server"><img src="https://agentmods.dev/badge/skills/cloudrift-ai/emmy/start-remote-server.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00061 | $0.02968 |
| Opus 5.5 | $0.00024 | $0.01187 |
| Sonnet 5.5 | $0.00012 | $0.00594 |
| Haiku 4.5 | $0.00006 | $0.00297 |
Grade C, and why
start-remote-server scanned grade C with 1 finding 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 1mo 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.
Reaches for credential fileshighPrivilege escalation
SSH keys, cloud credentials, git-credentials, .npmrc, /etc/shadow: reading these is how a config file becomes a credential leak.
[--ssh-key ~/.ssh/id_ed25519] \ How it starts
The opening of the file, as written. The whole thing — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Start Remote Server
Provision a fresh cloud GPU VM on the right provider (GCP or CloudRift) for the requested hardware, leave it running, and report back the SSH connection details so the user can run emmy deploy ssh ... or emmy bench --ssh ... against it later.
This skill does not deploy a recipe. To deploy + bench + teardown in one shot, use emmy deploy cloud or emmy bench directly — those handle their own VM lifecycle.
Preferred Path: emmy vm create gpu
For almost every request, use the GPU-name orchestrator command:
emmy vm create gpu \
--gpu "<full GPU name>" \
--gpu-count <N> \
[--provider cloudrift|gcp] \
[--name <prefix>] \
[--ssh-key ~/.ssh/id_ed25519] \
[--config config.yaml]
What this command does that the provider-specific commands do not:
- Resolves the GPU name to candidates from
emmy/hardware.py::GPU_INSTANCE_TYPESautomatically — you do not pass an instance type. - Iterates candidates in preference order (every
(provider, base)pair under the GPU key, all GCP zones fromGPU_GCP_ZONES). - Handles capacity fallback: on
CapacityExhausted(HTTP 5xx, 409, "no capacity", "out of stock", "Service Unavailable") it advances to the next candidate. On transient errors it retries the same candidate. - Cleans up orphans from failed attempts.
- Reads
config.yamlfor provider defaults (GCP image/disk/service account, CloudRift image override, etc.). - Stops on terminal errors (auth 401/403, malformed 400, unknown instance type 404) without burning the candidate list.
--ssh-key is the PRIVATE key path for this command (default ~/.ssh/id_ed25519). The orchestrator appends .pub internally and also uses the private key to verify SSH reachability. This is different from vm create cloudrift (see Manual Overrides below), which requires the .pub path explicitly. Do not pass .pub to vm create gpu.
--provider is binding. If the user named a provider, pass it. If the GPU isn't offered by that provider, the orchestrator raises ValueError with the available providers — surface that to the user; do not silently substitute.
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.
- 1mo ago First seen · 175 lines · 61 tokens per session scan C 3dcc3ae955cc
start-remote-server is a skill published in the GitHub repository cloudrift-ai/emmy (116 stars, last pushed today), licensed Apache-2.0. It adds 61 tokens to every session and 2,968 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it C with 1 finding (reaches for credential files). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
inference-aiops
Use this skill whenever the user needs to operate a GPU inference cluster — vLLM (OpenAI API + Prometheus /metrics) and Ray Serve / Ray Jobs (Ray dashboard), plus the single-process serving engines SGLang and TGI (Text Generation Inference): a one-shot cluster overview (deployments + total replicas + queue…
plano-agent-skills
Best practices for building agents and agentic applications with Plano, including configuration, routing, orchestration, guardrails, observability, and deployment.
nvcf-self-managed-cli
Install, operate, and tear down self-hosted NVIDIA Cloud Functions (NVCF) deployments with nvcf-cli. Use for control-plane or compute-plane install, status checks, cluster registration, function deploy/invoke, task create/list/cancel/delete, API keys, admin tokens, JWKS rotation, failed-install diagnosis, and…
nvcf-self-managed-prerequisite
Install the prerequisites the NVCA operator / compute plane needs before nvcf-nvca-install can succeed: the operator tool nvcf-cli (required by the compute-plane stack's register-cluster step), KAI Scheduler (for the KAIScheduler feature gate), and the SMB CSI driver (for the sharedStorage Samba sidecar PVCs). The two…
nvca-self-managed-install
Install or validate the NVCA Operator chart against a self-managed NVCF control plane from the native monorepo. Use when the control plane comes from deploy/stacks/self-managed and NVCA must be installed with stack-derived image repository settings.
nvcf-explore-stack
Navigate and explain the NVCF self-hosted stack inside the monorepo. Maps helmfile releases to their charts, image-source subtrees, helm hooks, namespaces, and needs: dependency chains. Reads deploy/stacks/self-managed/helmfile.d/.yaml.gotmpl and deploy/stacks/nvcf-compute-plane/helmfile.d/.yaml.gotmpl as the source…