The B2B twin of the Huxberry ads watchdog: a 35-point account audit on day one, then weekly checks on pacing, zero-converting audiences, bid efficiency — and LinkedIn's unique data on which job titles actually convert.
Weekly to the B2B marketing owner and the relevant BU lead.
Three of our four business units sell B2B, and none has any paid marketing motion. The playbook we adopted is natively B2B — its LinkedIn machinery transfers almost as-is. What's missing isn't design: it's a budget and an owner. Those are business decisions, not loop designs.
The 35-point baseline audit
Pacing, audiences, bids, demographics
Zero-converters and overpaying flagged
To the B2B owner + BU lead
Read-only by construction
For implementers and the technically curious. The full build sheet — verified queries, thresholds, and build notes — lives on the specs page.
| System | Role in this loop |
|---|---|
| LinkedIn Ads APIapproval takes days | The read-only data source — request access early; it's the documented bottleneck. |
| Playbook scriptsadopted reference repo | Battle-tested API scripts and the 35-item audit checklist — cloned at build time, adapted, never resold. |
| Agent Vaultcredential store | Where API keys live. Loops get read-only credentials from the vault; nothing is hardcoded in scripts. |
| Hermesagent runtime on our server | The scheduler that wakes the loop up. Each loop is a cron job under a Hermes profile; the planned bizops profile will host the business digests (IT loops run under vpsops). |
| Healthcheckshealthchecks.huxapps.com | The dead-man's switch. The loop pings it only after a clean run — if the loop dies or errors, the ping stops and Healthchecks raises the alarm independently. This is how 'never silent' is enforced by machinery, not promises. |
| Email renderer + gwsrender_email.py | All digests pass through one shared renderer: Huxberry-branded HTML, tables for repeated rows, a coral 'needs your response' box when the loop has questions, and an arrow link on every record. Sent from the loops mailbox via the Google Workspace CLI. |
| Model / brain | What it does here |
|---|---|
| None at run timedeterministic script | A normal cycle is a plain Python script — no AI tokens are spent unless a diagnosis or judgment step is actually needed. AI wrote the script; the script does the rounds. |
| GPT-5.5 via Codexthe bulk-work model | Wrote and maintains the mechanical parts — SQL, diffing, digest assembly. Effectively free on our existing subscription, so routine cycles cost almost nothing. |
| Claude Opusthe judgment model | Reviews alert wording, thresholds, and anything a human will read and act on. Post-Fable, Opus owns everything that ships. |
Account baselines from the day-one audit; weekly deltas against them.
Weekly email digest to the B2B owner + relevant BU lead.
Read-only credential; same construction as HUX-L3. No spend exists until humans create it — this loop just guarantees it's never unwatched.