Skip to main content

Create outputs with AI agents

The core of TrustedOSS is that an AI agent asks about your company's situation and automatically creates outputs that conform to the OpenChain standards. Instead of filling in blank templates yourself, you answer questions and the policy, process, and organization documents tailored to your company are generated. This page shows at a glance which agent creates what.

Agents at a glance

ChapterAgent (agents/…)Generated outputs
2 Organization02-organization-designerrole-definition, raci-matrix, appointment-template
3 Policy03-policy-generatoross-policy, license-allowlist
4 Process04-process-designerusage-approval, distribution-checklist, vulnerability-response, inquiry-response, (conditional) contribution / project-publication
5 SBOM creation05-sbom-guide, 05-sbom-analystSBOM(cdx.json), sbom-commands, license-report, copyleft-risk
5 SBOM management05-sbom-managementsbom-management-plan, sbom-sharing-template
5 Vulnerability05-vulnerability-analystcve-report, remediation-plan
6 Training06-training-managercurriculum, completion-tracker, resources
7 Certification07-conformance-preparergap-analysis, declaration-draft, submission-guide

You can see the actual form of the generated outputs in Policy Output Best Practices.

Which agent for which situation

  • If your goal is self-certification, run each agent in the order 2 Organization → 3 Policy → 4 Process → 5 Tools → 6 Training → 7 Certification. This is the required path.
  • If you only need a policy quickly, you can start from 03-policy-generator.
  • If you only want to check SBOM and vulnerabilities, use only the 5 Tools agents (05-sbom-*, 05-vulnerability-analyst).
  • If you need contribution or internal publication procedures, answer "yes" to the relevant question while running the 4 Process agent, and the conditional outputs are generated together.

Common way to run

All agents are run in the same way.

Check before running

First terminate the current Claude session (/exit or Ctrl+C), then run the command below in a new terminal. Put the agent name from the table above in place of XX-agent-name.

Bash
cd agents/XX-agent-name
claude

When the prompt opens, type start and answer the agent's questions. The generated outputs are saved in the output/ folder.

Next steps