Running the blueprint on Optimizely Agent Platform
The platform provides the mechanisms. The blueprint supplies what goes into them: what the skills say, who owns each agent, what the quality bar is, and which number the workflow must move.
Blueprint to platform
| Blueprint element | Agent Platform capability | What we add |
|---|---|---|
| Agent Roster | Simple assistants, specialized agents, workflow and autonomous agents; Agent Directory | A stage-by-stage decision: vendor agent, custom build, or orchestrated |
| Process Map and orchestration | Workflow agents with triggers (chat command, schedule, webhook, email), conditions, loops and code steps; Content Marketing Platform webhooks can trigger workflow agents | The map that decides where triggers, branches and loops belong before anything is built |
| Knowledge Layer | Organization and personal skills (brand voice, compliance rules, templates, reference files); Brand Packs as a single source of truth for brand identity | The content and ownership of that layer |
| Governance Gate | Quality tab (Output Evaluation, Execution Guardrails, Execution Advisor); evaluation agents inside workflows | Thresholds, mandatory checks, and the line between agent-led and human-controlled decisions |
| Governance Model | Agent Activity Logs: status, who triggered a run, approximate credits, evaluation score | A named human owner per agent and a review routine |
| KPI Tree | Per-run credits and evaluation scores in the logs | A day-one number and stage metrics, including cost per outcome |
| Training Track | Optimizely Academy learning path; admin, builder and user documentation | A role-based certification path and handoff |
Sources
Capability statements are from Optimizely's Agent Platform documentation.
Ready to pick a pilot?
- 1 Choose a pilot
- 2 Run the diagnostic
- 3 Ship a quick win