Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/synthetic-sciences/openscience/modalnpx skills add synthetic-sciences/openscience --skill modalgit clone --depth 1 https://github.com/synthetic-sciences/openscienceWrote 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/synthetic-sciences/openscience/modal)<a href="https://agentmods.dev/skills/synthetic-sciences/openscience/modal"><img src="https://agentmods.dev/badge/skills/synthetic-sciences/openscience/modal.svg" alt="Measured on agentmods" 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 | $0.00054 | $0.00756 |
| Opus 5 | $0.00027 | $0.00378 |
| Sonnet 5 | $0.00011 | $0.00151 |
| Haiku 4.5 | $0.00005 | $0.00076 |
Grade A, and why
modal-serverless-gpu 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 today.
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 — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Modal through compute_job
Modal is an OpenScience compute target, not an agent-controlled SDK. Prepare ordinary workspace files and call compute_job with target { kind: "modal" }. The JobBroker owns credentials, the reviewed dispatch card, sandbox lifecycle, logs, cancellation, recovery, and output delivery.
Contract
- Read
<compute-capability>for the configured/disabled/unconfigured state. Do not probe availability by launching work. - Never inspect Modal credentials or
~/.modal.toml; never install or invoke the Modal Python SDK or CLI. - Commands are ordinary shell commands inside
/workspace, such aspython analysis.py. - The
packageslist builds Python dependencies into the reviewed image before execution. Pin important versions. - With no
cwd, omitteduploadsstages safe ordinary files from session scratch. With acwd, it stages from that directory. Pass explicit globs to narrow inputs; passuploads: []only when the job needs no workspace files. - Declare every output needed after the run in
artifacts. Undeclared files are not delivered. - Use GPU
nonefor CPU work. Choose GPU count, CPU, memory, and timeout from the workload; do not invent price or duration guarantees. - Network
noneis the safe default. Use unrestricted network only when the command genuinely needs remote access and the approval card shows it. - Refer to stored secrets only by symbolic name. Never place secret values in commands, files, prompts, logs, or plans.
- The
compute_jobcard is the authorization boundary. Do not request a second prose approval or treat chat text as dispatch authorization.
Workflow
- Prepare a reproducible script and small inputs in session scratch.
- Call
compute_jobwithaction: "plan"when parameters still need inspection, otherwiseaction: "start". - Include a clear name, purpose, command, target, packages, resources, upload selection, and artifact globs.
- After approval, use the returned job ID for status, logs, artifacts, cancellation, delivery retry, or release. Do not create a duplicate job merely because a long job is still running.
- Report only states and outputs returned by
compute_job. Preserve failures and uncertainty.
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.
- today First seen · 54 lines · 54 tokens per session scan A 5b61a97ac609
modal-serverless-gpu is a skill published in the GitHub repository synthetic-sciences/openscience (3,432 stars, last pushed today), licensed Apache-2.0. It adds 54 tokens to every session and 756 once invoked, about $0.0003 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 2026-09-03.
Other skills, from other repositories
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.
modal-serverless-gpu
Serverless GPU cloud platform for running ML workloads. Use when you need on-demand GPU access without infrastructure management, deploying ML models as APIs, or running batch jobs with automatic scaling.