Channels, budget, capacity, privacy rules, and voice examples
Operating guide
Know when to run it and who leads
Lead Bot
Pulse
Typical first run
15–30 minutes for a first manual run
When to run it
When a person supplies the approved brief and starts the workflow
Use this workflow when you need a focused growth test with a verified baseline, one primary metric, and draft materials ready for review.
Access for the first run
The approved inputs listed on this page
Only the files, sources, and services required for this run
No sending, publishing, spending, or live-system changes without approval
Run the first version one step at a time
Give one Bot the complete job
Use one Bot for the first run. Add a specialist only when the job becomes complicated enough to need a separate role, different access, or independent review.
1
Pulse
Review the approved evidence, verify the baseline, design one measurable experiment, and draft the smallest useful variants
Produces: An experiment plan with one primary metric, decision rules, evidence gaps, and two draft variants
Use the approved inputs I provide. Review the approved evidence, verify the baseline, design one measurable experiment, and draft the smallest useful variants. Return this result: An experiment plan with one primary metric, decision rules, evidence gaps, and two draft variants. Ask me about missing information before you continue.
Overall request
Design one growth experiment for this approved goal. Verify the baseline, use one primary metric, set success and stop rules before the test, and draft two variants without changing a live campaign.
Use this as the complete first-run request.
Decisions for a person
Approve the problem interpretation, metric, and decision rules
Approve targeting, budget, outreach, and any tracking change
Approve the variants and decide whether to launch, revise, or stop
First test
Try the workflow with low-risk material
Use one approved historical data export to design a test and draft two variants without changing a live campaign.
Check the result
Check the result against your original material. You make the decisions listed below. Use one approved historical data export to design a test and draft two variants without changing a live campaign.
If the workflow stalls
If the result is incomplete, mark the missing information and ask the same Bot to correct it. Add another Bot only when a genuinely separate role or independent check would help.
Set it up in Hermes Desktop
Begin with one Bot and one manual run
Open the linked starter page, download its .tar.gz profile, and import it from the Profiles screen in Hermes Desktop.
Review the profile’s SOUL.md, Bot Passport, and requested access.
Run the complete request in that Bot’s chat with low-risk material and review the result.
Add a schedule, outside connection, or optional specialist only after the manual version works.
