This page helps Genesis Mission RFA teams choose and safely adopt model access, coding assistants, scientific-agent frameworks, and the infrastructure that supports them.
The intended audience includes domain scientists, research software engineers, data scientists, workflow developers, and technical leads. You do not need to build a multi-agent system to benefit from AI.
Start with the lowest level of autonomy that solves the problem. A direct model call, a conventional script, or a coding assistant is often a better first choice than an autonomous agent.
Before you connect project data or tools
Model access and agent access do not, by themselves, authorize a data use. Review the companion pages before enabling data access or tools:
Choose your starting point
| What you need to do | Start with | When things get more complex ... |
|---|---|---|
| Ask questions, summarize approved material, extract structure, or test a prompt | A direct Model Access Gateway (MAG) call or approved chat interface | Switch to an agentic system like those described below |
| Write, explain, refactor, or test code in one repository | One coding assistant: Claude Code, Codex CLI, or OpenCode | Utilize agents designed for reusable scientific planning, specialist tools, remote resources, or persistent coordination like those shown below |
| Capture project conventions and a repeatable procedure | AGENTS.md, CLAUDE.md, and/or a task-specific skill file | Utilize structured agentic systems that can manage complex workflow, invoke tools autonomously, or coordinate multiple components |
| Build hypothesize-plan-execute scientific workflows | URSA | Add specialist agents via MCP (e.g. MADA below) or adapt for complex deployment with tools like Academy |
| Setup, manage, and analyze simulation workflows | MADA with MADA Tools | Pair with general purpose scientific workflow agents like URSA |
| Build stateful, asynchronous, long-running, or federated agents | Academy | Use Academy's deployment patterns and pair it with the needed model, tool, data, and compute services |
| Move data, execute functions remotely, or connect transfer and compute steps | Globus Transfer, Compute, and Flows | Pair Globus with the agent or workflow layer that makes the scientific decisions |
| Serve a self-hosted model across an HPC allocation | ExaServe | Use only when the team has a supported site, an allocation, model-serving expertise, and an authentication/rate-limiting plan |
| Run a stable, deterministic transformation or simulation pipeline | A conventional script, notebook, workflow engine, or scheduler | Add an agent only for steps that genuinely require interpretation, adaptation, or tool selection |
Building a practical architecture pattern
Building a practical architecture pattern is all about finding the right tools. Here, we provide helpful starting resources for making choices on three classes of tools:
The tools can fill different niches in a single system. A coding assistant can be used to build, edit, and test AI code. Academy can be used to construct and host stateful agents across resources. Globus exposes access cross-site resources. The URSA library can be used to construct a local workflow that leverages these agents and resources and interface with the MADA library for simulation. URSA and MADA would leverage the MAG for tokens and access additional MCP tools as necessary.
This is an example; not necessarily a requirement. Use only the layers the scientific objective needs.
Documentation
As with any workflow, AI workflows need documentation. Documentation allows your workflows, or workflow outputs, to be reusable, shared, evaluated or deployed.
Here we provide a variety of templates and resource pages to get your started with documentation for AI and Agents:
Agents and Skills guidance: Documenting repository behavior and repeatable procedures.
Agent card template: Documenting capabilities, interfaces, authentication, tools, side effects, runtime, memory, intended use, limitations, risk, evaluation, and human oversight.
Model card template: Documenting underlying model details, data, evaluation, limitations, and governance.
A reusable agent should not be considered ready merely because it completes a demonstration. It should have an owner, version, documented permissions, evaluation evidence, known limitations, and a defined support and retirement path. In the same way that data workflows benefit from a data management plan (DMP), models and agents also function best with an model management plan (MMP).
A first successful hour
1. Request access to the Model Access Gateway
Navigate to Genesis Mission Resource Request Page and request access to resources you need for your RFA team, including the Model Access Gateway.
2. Confirm access and allowed data
Obtain an American Science Cloud account.
Follow the Model Access Gateway documentation to request access, create a project credential, identify the correct endpoint, and list the model IDs available to your project.
Do not copy an example model ID from a reference card without checking the current model list for your credential.
Confirm which project data may be used before sending the first prompt.
3. Work in a safe project copy
Create a branch, clean clone, container, or disposable workspace. Confirm that you can restore the original state.
4. Select one coding assistant
Teams new to agentic coding should standardize on one tool initially. Choose based on the team's existing model/provider preferences and workflow needs, not on a blanket claim that one assistant is always best.
See Choosing a Coding Assistant for reference cards and additional details.
5. Add project instructions
Use Developing AGENT, Claude, and Skill Files to create a small, testable context file. A useful starter structure is:
# Project purpose
What scientific question or software objective does this repository support?
# Approved inputs
Which files, datasets, endpoints, and data classes may the assistant use?
# Allowed actions
What may it read, create, edit, execute, or call?
# Prohibited actions
What must it never access, change, delete, transmit, publish, or submit?
# Commands
How should it install dependencies, run tests, lint code, and reproduce results?
# Scientific acceptance criteria
What quantitative checks, physical constraints, baselines, or expert reviews define success?
# Artifact locations
Where should generated code, logs, plots, reports, and temporary files go?
Prefer a vendor-neutral AGENTS.md when possible. Add CLAUDE.md only for Claude-specific behavior and a skill file for a narrow, reusable procedure.
6. Run a read-only task first
A safe first prompt is:
Read-only task. Do not modify files, execute commands, access the network, or call external services.
Explain the repository structure, identify the main scientific workflow, list assumptions you are making, and propose three small tasks that can be validated with existing tests or data.
Then allow one small change with explicit files, commands, and acceptance tests. Review the diff and test output before increasing autonomy.
7. Add an agent framework only when the requirement is clear
Write down why a coding assistant, direct model call, or conventional workflow is insufficient. Identify required tools, state, permissions, deployment targets, approval points, failure recovery, and evaluation criteria before selecting URSA, MADA, or Academy.
Troubleshooting and support
Access or model errors: Recheck the current MAG documentation, credential environment variable, endpoint, and model list. Model availability is project-specific.
Unexpected agent behavior: Stop the run, preserve logs and artifacts, inspect the exact tool calls and file changes, restore the workspace if needed, and reduce permissions before retrying.
Scientific uncertainty: Pause automation and involve the domain lead. Do not use model confidence or fluent prose as evidence of correctness.
Security concern: Revoke exposed credentials, stop affected services or jobs when safe to do so, preserve relevant evidence, and follow your institution's security-incident process.
Repository content issue: Open an issue in the RFA launch repository with the page, tool version, operating environment, observed behavior, and non-sensitive error details.
Additional Resources
Several additional resources have been made publically available to help get you started with AI and Agentic workflows.
Explore additional resources here!
Submit a support ticket for any of the following:
Connect your RFA team with services like data management planning, supercharging your scientific workflows with AI best practices, and cross-cutting AI capabilities.
Have an issue, suggestion, or addition to this content.