Write the method once. Run it when the work returns.
Describe a repeating brief in plain language. Pin it to a project and the skills it should use, then run it on a schedule or when someone asks.
Waiting
Health / Access practiceInternal
Project Solace pipelineClient
NICE this weekPublicBrief
- NICE published two appraisals that sit on Project Solace's list.1
- The consultation on CG189 is still open, so treat that line as provisional.2
- Raise both with the access lead this week.3
A method the firm can run again.
Each run is the same research loop: sources attached, claims marked, ready for an expert to review.
A project and the skills it should use.
Pin it to a client engagement, or leave it on internal material. Attach the skills the firm already uses for tone, method, and checks.
- A client project, or organisation knowledge
- Skills for tone, method, and checks
- Sources stay the ones that project and those accounts already allow
Health / Access practiceProject Solace pipelineNICE this weekNo project or skills attached yet.
On a Tuesday, or when someone asks.
Give it a cadence, or keep it as a button the team presses. A finished run is a conversation you can open and review.
- Daily, weekly, or monthly, at an hour you set
- Run now from the agent, without waiting for the clock
- Pause it without deleting the method
Scheduled
Tue 07:30
Weekly
Brief
Project Solace - what NICE published
On demand
Brief
Project Solace - what NICE published
Same brief. Two ways to start it.
Describe the work. Hebno writes the agent.
Say what you want done. Hebno turns that into a named agent, with a schedule if you asked for one.
- Plain language, not a workflow builder
- A name, a brief, and a trigger come back from the sentence
- You still choose the project, the skills, and who can run it
- Pull new NICE and MHRA items that touch Project Solace's pipeline.
- Mark anything still in consultation as provisional.
When the same work keeps coming back.
Hebno suggests an agent for the brief the team already repeats.
- Suggestions come from how the team actually works
- Accept a suggestion and it becomes an agent you can edit
- Ignore it and the work stays in chat, unchanged
From a sentence to a run you can review.
- 1
Describe
Tell Hebno the work, in plain language. It becomes a named agent.
- 2
Pin
Pin a project or firm knowledge, attach skills, and choose the sources it may use.
- 3
Run
Leave it on a schedule, or press run. Each run is a research turn with sources attached.
- 4
Review
Open the conversation, check the claims, and put a name on it before a client sees it.
What sits on every agent.
Personal or organisation
Keep one to yourself, or share it. Owners and admins decide who can run an organisation agent.
Each run is a research turn
The output is a sourced draft, not a log. Open it in chat and review it like any other brief.
Skills travel with the method
The same instruction packs you attach in chat can sit on the agent, so tone and checks do not drift.
Pause without losing it
Stop the schedule. The method and its run history stay.
Nothing leaves without a person
A run waits in the workspace. An expert still reviews it.
Visible in spend
Runs use the models you allow and count against the team and project you already meter.
One loop, not a pile of tools.
Research, drafting, sources, and cost sit in the same workspace, so work does not have to be moved between them.
- AgentsWrite the method once. Run it when the work returns.You're here
See a method become a Tuesday morning brief.
Book 20 minutes and watch a team go from brief to source-backed draft, with every source visible and ready to review.