THE KITCHEN · unstress health
03 · The engine · formerly unlisted — promoted to the front door · v0.2

The room where the machine lives.

Door 05 shows the effect on the week; this room shows the shape underneath. Eighteen automations, arranged by where they live in the production cycle and by how soon they can ship. Depth on tap; nothing hidden.

Audience honesty. This door started life unlisted, and earned its spot on the front of the room — the progress inside was too concrete to keep behind a whisper. It's here for whoever wants to kick the tyres — the tyre-kicker's likely Sophie, doing the plumbing for the new site while the rest of the room stays feelings-first. Ron is welcome, of course. So is future-Nick, who'll be reading this to remember what he decided.

The shape, in one screen

Three beats. The machine tends the middle one.

All knowledge work runs the same loop — gather, decide, learn. The engine takes the routine parts of the middle and keeps the routine off Ron's desk. Recording, judgement, taste — all of it stays Ron's.

Beat 01 · gather
Before the record

Guest research, archive cross-refs, a one-page brief in the inbox 48 hours before Ron sits down. Ron reads, annotates, arrives at depth.

Beat 02 · decide & act
After the record

Transcript, show notes, blog draft, captions, thumbnail brief, clips, nurture — a single production document that Ron reviews once. Approve, edit, redirect. Nothing sends without sign-off.

Beat 03 · learn
After the send

Ron's edits become the calibration data — the machine gets a little more Ron with every episode. The archive deepens; every future dossier stands on more ground.

One sentence of context, then no more trend talk on this page. Anthropic's own numbers, June/July 2026: more than 90% of Cowork activity is non-coding, and business operations plus content creation account for roughly half of all usage. Ron's use case is the tool's mainstream, not an experiment.

One recording in. The rest arrives on its own rhythm.

Read left to right. Phases are temporal — the sequence a single episode moves through. Each cell is coloured by tier: what ships now, what waits on a dependency Ron controls, what lives on the horizon. The empty cell in the middle is deliberate — the recording itself stays Ron's.

T1 — ships now
T1 · conditional on the GHL migration existing
T2 — needs a build
T3 — six months out or further
Trust-critical checkpoint · Ron-owned, non-skippable
Pre-production
Brief in the inbox 48 hours before Wednesday.
01Research Dossier
04Guest Curator
05Calendly → Brief
14Corporate Prospect DB
During recording
Ron records. Riverside captures. Nothing else.
n/a — the recording is the point.

Automating presence is a failure mode. Riverside is an upstream input, not an automation target.

Post-production
Within hours, a near-finished production document is waiting.
09Transcript Pipeline
02Production Package
08Thumbnail Brief
06Voice Model Refinement
15Second-Book Research
Post-publication
Every platform gets its shape. Ron keeps his Thursday.
03Distribution Package
10Reels & Shorts
07Nurture Sequencer
11Membership Router
16Healthy Bite Repurposing
Compounding
The archive deepens. Each episode makes the next one sharper.
12Coaching Brief
13Archive Ingestion
17Archive Knowledge Base
18Speaker Prospect Enrichment

What changed since April. Nine of the eighteen now ship in the first six weeks — 04 (Guest Curator) and 06 (Voice Model) moved from T2 into T1 because Cowork scheduled tasks shipped on the 7th of July, and because a working preference-compounding pattern went public a fortnight before that. Three more (05, 07, 11) become T1 the moment Sophie's GHL migration lands. Details on every card.

Each one, at the depth its tier warrants.

Click a card to expand. T1 entries are specified at working depth. T2 specifies the bridges T1 outputs rely on. T3 is sketched — enough to see the shape, not enough to over-engineer what lives six months out.

T1 · ships now

Nine automations. These define whether the engagement lands.

Seven from the April tier, plus two the July capability shift promoted in.

01Research Dossier
Pre-production
TriggerBooking confirmed; the dossier lands in Ron's inbox 48 hours pre-record. Scheduled task, no window to open.
OutputsPer-guest brief mapped to Ron's category taxonomy, with points of tension & agreement and three suggested threads. Archive cross-refs by episode title only until transcripts land.
CheckpointRon reads, annotates, adds or removes threads. The dossier is a starting point, not a script.
DisplacesRon's Tuesday Perplexity session. No impact on Amy's workflow — this is Ron's own prep, net-new capacity.
Failure modesHallucinated guest claims (cite URLs); stale profile (timestamp sources). The two dossiers already in the room are the ground truth for how this reads.
02Production PackageTrust △
Post-production
TriggerEdited transcript available (from 09 or manual deposit).
OutputsA single production document: edited transcript, show notes, five takeaways, description, keywords, hashtags, top categories, blog post, pull quotes, guest bio, YouTube timestamp chapters. Plus a JSON sidecar for 03.
CheckpointRon reviews the full package in one session — the Thursday 45 minutes. No auto-publish, ever at this stage, until at least a dozen calibration episodes and Ron opts in.
DisplacesThe largest single reclamation of hours in the pipeline: production notes, blog prompt, transcript cleanup. Amy's role shifts — the drudgery is lifted, the ground-truth workflow doc stays hers.
VoiceOne voice model, calibrated every time Ron edits. Drift protection: a rolling 25% edit threshold over three consecutive episodes retrains.
03Distribution Package
Post-publication
TriggerRon approves the Production Package (02 checkpoint event).
OutputsPer-platform drafts: WordPress / GHL, Omny description with standard footer, YouTube with chapters, Spotify, LinkedIn, Instagram, Facebook, TikTok, Lnk.Bio, Metricool schedule.
CheckpointPer-platform draft review, or standing rule — Ron's choice per platform, after four clean runs. WordPress / GHL and Lnk.Bio stay hard-draft regardless.
Distribution honestyRon told us LinkedIn is where the "serious connections" are — it landed Hilton APAC — and YouTube's 9,000+ organic subscribers came without marketing. Instagram and TikTok are weighted lower on this page for the reason Ron gave: the audience there is "flaky." His words, not ours.
DependenciesBridges the current stack until Sophie's GHL migration lands, then routes through it. 02 approval blocking.
04Guest CuratorTrust △
Pre-production
Why it moved upCowork scheduled tasks (shipped 7 July) mean the weekly Friday run happens on its own — no window to open. That was the whole reason this was T2.
TriggerWeekly Friday morning — produces Ron's Friday shortlist.
Outputs5–8 candidates per thin category, each with recent work summary, LinkedIn connection path, framing rationale, draft outreach note.
CheckpointRon approves every candidate individually. No bulk approve. No outreach sent in his name without per-candidate sign-off.
Failure modesHomogenous slate — minimum diversity across geography, discipline, stage, consensus position.
06Voice Model Refinement Loop
Post-production
Why it moved upThe pattern that makes this cheap — derive rules from Ron's edits, append to a voice-rules file every generator reads — went live in an MIT-licensed repo in June. Ron's edits are the calibration data; that's the whole shape.
TriggerEvery approval event on 02. Asynchronous — does not block production.
OutputsVersioned voice model + plain-language drift report to Ron ("I noticed you replaced 'explore' with 'unpack' three times — adding this to the model").
CheckpointRon can roll back any model version. Maintains a "voice locked" phrase list that cannot be retrained away.
RoleThe single voice-calibration point. Shared surface for every generation step in the pipeline.
08Thumbnail Brief Generator
Post-production
TriggerApproved Production Package. Runs in parallel with 03.
OutputsComposition brief + draft preview images for three variants: Omny (square), WordPress (16:9), YouTube (16:9 with hook text).
CheckpointRon approves brief; an editor can finish in Canva from the brief in a transition role. Human craft is preserved on purpose.
DisplacesAround 45 minutes an episode of thumbnail production. Partial only — the brief is automated, the asset stays with an editor.
09Transcript Pipeline
Post-production + retroactive archive
TriggerOngoing: raw recording deposited. One-time retroactive: batch run across Riverside, Drive archive, YouTube backfill.
OutputsStandardised text per episode + JSON sidecar with timestamps, in the Drive transcript folder. Registered in 13.
Cost envelopeUnder $300 for the full retroactive corpus, roughly nine hours of machine time. Cheaper than two weeks of a VA.
DependenciesRiverside access (blocking for Phase 1). Two Drive folder 404s (2021 and 2023). Whisper API key or local compute.
DisplacesThe manual transcript pass on new episodes. The retroactive archive is net-new — no displacement of Amy's work; it's work she has not been asked to do.
10Reels & Shorts Generator
Post-publication
Riverside honestyRiverside already auto-generates magic clips — Ron said the 16:9 ones are "terrific" and the 9:16 ones don't work so well. This automation routes and places what Riverside makes; it does not claim to create clips Riverside already creates.
TriggerApproved Production Package + transcript from 09.
Outputs3–5 clip proposals per episode with start/end timestamps, platform-specific captions, thumbnail text. Handed to the Riverside clipping engine.
CheckpointRon drops any of the 3–5 clips from queue. Auto-publish governed by 03's per-platform rule.
VoiceClips are Ron's actual speech — voice preservation is inherent, not generated.
12Coaching Brief (pre-call)
Compounding
TriggerProspect books Calendly + completes intake form (webhook).
OutputsOne-page pre-call brief: intake mapped to the five pillars and the category taxonomy, a few relevant episodes with rationale, self-assessment result, a suggested conversational entry point.
CheckpointBrief is advisory. The conversation is Ron's. The prospect never sees automated output — hard trust line.
DisplacesAd-hoc intake review. Net-new — no impact on Amy.
Failure modesOver-confident clinical inference — hard rule: no pathology-naming, only framework-mapping.
T1 · conditional

Three that ship the moment Sophie's GHL migration lands.

GoHighLevel released an official Model Context Protocol server in June — free, five-minute token paste, roughly forty tools growing to several hundred. The whole reason these three were T2 in April was "GHL integration = architectural build." That build is now plumbing.

The dependency is Ron-side, not capability-side. Sophie's simpler-site build on GHL is the piece that needs to exist before these three can run. Nothing here replaces that build; the MCP connection sits underneath her site, not next to it. Supported, not replaced.
05Calendly → Brief, full pipeline
Pre-production · coaching surface
TriggerCalendly booking + intake. Hardened extension of 12.
OutputsTrack-routed brief (individual vs corporate). Corporate uses Hilton-style references; individual uses framework mapping.
CheckpointAs 12. Plus: Ron approves track assignment for ambiguous cases.
DependenciesGHL migration; payment-confirmed gate; 12 operational.
07Nurture SequencerTrust △
Post-publication
TriggerPost-episode (03), post-call (12/05), post-intake (05 submission).
OutputsSegmented email sequences in GHL. Individual and corporate tracks with distinct copy. Multi-touch cadence: episode → day 3 → day 7 → day 14.
CheckpointPre-send approval by default. Segment-level auto-send after four clean runs, not campaign-level. Corporate track stays manual for the first ten sends.
VoiceRon's voice at full strength on framing paragraphs. Register closer to a note than a newsletter.
DependenciesGHL migration; list hygiene reconciliation at migration cutover.
11Membership Content Router
Post-publication
Trigger03 publishes + Ron flags member-priority (or category-auto-route).
OutputsMember-area entries with pillar / category navigation. Supplementary content drafts (deeper takeaways, framework excerpts, related self-assessments).
CheckpointRon approves supplementary content before publication. Over-gating protection: Ron's explicit gating choice.
DependenciesGHL migration (MemberPress currently brittle); 02 operational.
T2 · needs a build

One left in this tier after the July shift.

Everything else in the original T2 either moved up (04, 06) or became conditional-T1 (05, 07, 11). This one stays where it was because it earns its complexity honestly.

13Archive Ingestion & Tagging
Compounding
TriggerEvery 02 approval; one-time batch for the retroactive transcript corpus.
OutputsVector index + structured metadata. Tagged by category, pillar, guest, theme, with quote-level granularity.
CheckpointNone — ingestion is not user-facing. Retrieval quality is Ron-facing via 17.
Failure modesTagging drift over time — periodic re-indexing against a gold-standard set.
Dependencies09 Transcript Pipeline for the corpus.
T3 · horizon

Five, sketched — not proposed for building soon.

Their purpose on this page is to show how T1 outputs compound into T3 inputs — not to over-engineer what lives six months out. Capacity restoration first; ambition later.

14Corporate Speaking Prospect DB
Pre-production · corporate track
ShapeNightly LinkedIn signal accumulation; fit-scoring against Ron's offering; warm-path discovery via second-degree.
CheckpointRon approves outreach per-prospect (same pattern as 04). No bulk, no automated outreach in his name.
ContextRon has explicitly recoiled from scale (declined Hilton-sized corporate pursuit). This automation systematises the Hilton outcome without becoming a growth-ambition engine.
15Second-Book Research Extraction
Post-production · writing surface
ShapeRon-invoked query against the corpus. Surfaces quotes, episode references, framework passages scoped to a chapter or theme.
CheckpointThe writing is Ron's. This is research support, not co-authorship.
Also feedsRon's own e-book-with-embedded-video idea — clips from Riverside's 16:9 pool, placed inline against text so a reader can watch the passage they just read. His idea, on the page here because it hasn't been added to the register yet.
Dependencies09, 13, 17. Impossible without the transcript pipeline.
16Healthy Bite Repurposing Engine
Post-publication
ShapeRe-packages existing Drive assets into Healthy Bite solo-commentary episodes. Draft scripts in Ron's voice based on archive cross-references.
Hard constraintNo synthetic voice, ever. Ron records; no text-to-speech of Ron anywhere.
ReactivatesThe unused Q&A clips, guest reel folders, Pillars clips and Sofie promos already sitting in Drive.
17Archive Knowledge Base
Compounding
ShapeQuery-able on Ron's full body of work. Citation-grounded retrieval (episode + timestamp + quote). Never publishes anything itself — private-facing.
CheckpointRon reviews answers. The KB is a reference tool.
Failure protectionNo summary claim without a retrieval quote — citation-required rule prevents confident hallucination.
18Speaker Prospect Enrichment
Compounding
ShapeGenerates speaking-proposal drafts with archive evidence, ROI references, Hilton-precedent tailoring.
CheckpointRon approves every proposal before send — high-stakes external artefact.
DisplacesThe speaking-proposal effort that currently makes Ron decline inbound invitations.

Thursday. Forty-five minutes. A stack of cards.

The week's promise (door 05) was "one review session, one coffee. Approve — it goes." This is the shape underneath it. Every item that would carry Ron's name arrives as a card with a proposed action. Ron talks his way through the pile — approve, edit, redirect, drop — and nothing sends without his sign-off.

The card-sweep, in a sentence: every artefact that leaves this room passes through Ron's eye first.

Some cards are quick — a thumbnail brief, a clip shortlist. Some deserve real thought — a nurture email to a corporate prospect, a guest outreach note in Ron's name. The three checkpoints below are not procedural. They're the load-bearing walls. Remove any one and the trust structure collapses. So they don't get removed.

Checkpoint 01

Guest selection— 04

The podcast is Ron's voice, curated by Ron. If the machine recommends and automates outreach, the podcast gradually becomes an artefact curated by the machine. That's the wrong direction. What makes this show work at hundreds of episodes is that every guest is a guest Ron chose.

  • Bulk approve is not offered — every candidate, separate click
  • No outreach in Ron's name without per-candidate sign-off
  • Draft outreach is advisory; Ron can replace it
  • Persistent "do not recommend" list Ron maintains
  • Volume calibrated to Ron's attention, not the machine's throughput
Checkpoint 02

Pre-publish show notes— 02

This is the voice surface. If the package publishes without Ron's review, every downstream distribution piece carries content Ron did not sanction. Voice drift becomes voice replacement — the slowest and most damaging failure in the pipeline.

  • 30–45 minutes per episode; design target, not nice-to-have
  • No auto-publish at this stage, ever, until at least a dozen calibration episodes and Ron opts in
  • Drift surfaced in plain language, not a silent metric
  • Voice model version-controlled; Ron can roll back
  • Ron maintains a "voice locked" phrase list that cannot be retrained away
Checkpoint 03

Pre-send nurture— 07

Email lands in inboxes with Ron's name on it. A tone-wrong email to a coaching prospect degrades the relationship; to a corporate prospect, costs the pipeline; to a subscriber, triggers an unsubscribe. Unlike a bad draft (which Ron reviews), an automated send lands in people's lives without his eyes on it.

  • Pre-send approval by default
  • Segment-level auto-send after four clean runs — not campaign-level
  • Corporate track stays manual-approve for the first ten sends
  • Frequency-cap logic per subscriber prevents campaign stacking
  • Diff-log every send — Ron can see what changed even when auto-approved

All three checkpoints sit at the boundary where Ron's voice reaches the world. Guest selection reaches his audience through who he platforms. Show notes reach them through what he appears to have said. Nurture reaches them through what he appears to have written. Protecting those surfaces is not a feature — it's the engagement.

The compounding, traced in four chains.

Static tier separation misses the thesis. The work compounds — each T1 artefact deepens what a T2 automation draws from, which deepens the corpus a T3 automation serves. Four chains trace this explicitly. Two are running-shape today; two are horizon.

Chain AContent
The dossier is the entry point of an engine that compounds. Every one Ron reads makes the next slate of guests sharper.
T1 · pre-production
01 Research Dossier
Per-guest brief; taxonomy mapping; archive cross-refs.
T1 · pre-production
04 Guest Curator
Taxonomy enriches from dossier cross-refs; emergent themes surface.
T3 · compounding
18 Speaker Prospect Enrichment
Proposals drawn from enriched taxonomy; cited via 17.
T3 · pre-production
14 Corporate Prospect DB
Warm-path second-degree discovery.
Chain BCorpus & writing
Podcasting and writing — the two things Ron said he really enjoys doing. Three books already on the go; the Mastering Oral Health course being reflowed into chapters via Claude. This machine feeds that work; it doesn't promise a book pipeline that isn't built.
T1 · post-production
09 Transcript Pipeline
The whole archive, made searchable.
T1 · post-production
02 Production Package
Per-episode package; 06 captures the edits.
T2 · compounding
13 Archive Ingestion
Indexed corpus, quote-level granularity.
T3 · compounding
17 Archive Knowledge Base
Cited retrieval on Ron's full body of work.
T3 · post-production
15 Second-Book Research
The book writes from a corpus, not from memory. Also the vehicle for Ron's e-book-with-embedded-video idea.

What this chain doesn't promise. A book pipeline that writes chapters for Ron. The course-to-book chain Ron already runs in Claude — that's what "already" means; this machine feeds it, and its own e-book-with-embedded-clips shape is his idea, sitting here waiting for a build slot, not shipping in the first six weeks.

Chain CCoaching & corporate
The chain that turns Ron's offer from "we did that once with Hilton" into a repeatable shape.
T1 · compounding
12 Coaching Brief
Per-prospect brief against intake + archive.
T1c · pre-production
05 Calendly → Brief (full)
Track routing; corporate intake shape distinct.
T1c · post-publication
07 Nurture Sequencer (corporate)
Hilton references; Deloitte $1 → $5.60 arithmetic; legislation framing.
T3 · pre-production
14 Corporate Prospect DB
Warm-path second-degree discovery.
T3 · compounding
18 Speaker Prospect Enrichment
Tailored proposals via 17.
Chain DDistribution
Weighted the way Ron already weights it — LinkedIn and YouTube first, Instagram and TikTok second. Plus the shift he named himself: showing up in ChatGPT's answers, not just Google's.
T1 · post-publication
03 Distribution Package
Per-platform packaging; LinkedIn + YouTube weighted first.
T1 · post-publication
10 Reels & Shorts
Clip selection over Riverside's 16:9 pool; captions attenuated per platform.
T1c · post-publication
11 Membership + 07 Nurture
Member deepening; subscriber and corporate nurture.
T3 · post-publication
16 Healthy Bite Repurposing
The Drive's unused clip library reactivated into evergreen solo content.

AEO — answering-engine optimisation. Ron's marketing friend's thesis, in Ron's own words: everyone is still doing Google organic, but ChatGPT is where the attention is going. The corpus that 09 and 13 build is what makes Ron's answers findable there. Feeds Chain D at the horizon, not the first six weeks.

Amy is the spine of this room, not a target of it. The workflow document she keeps — the one Ron said she updated for the discovery call, at his request — is the artefact every automation on this page is scored against. If the machine hasn't lifted a piece of work she carries today, and hasn't lifted it in a way she can see and audit, it isn't shipping. Nothing here displaces her work; the whole point is to move the drudgery off her hands and leave the judgment on them.

Two, sitting in plain view.

The first has been true since the mockup. The second showed up on July 7th, with a Cowork release. Both stay named — that's the register the whole engagement runs in.

Ron wants less management and more control at the same time.

Those two pull against each other in every non-trivial automation. Visible in the per-episode review, in guest curation, in every nurture send. The architecture doesn't resolve the tension by compromising between the poles. It resolves it by allocating — category by category, artefact by artefact.

Less management wants

Fewer decisions per week. Fewer logins. Fewer approvals. Fewer edge-case escalations. The hours of weekly plumbing collapsed into a shape that fits in Ron's pocket.

More in control wants

Ron's voice stays Ron's voice. The podcast he puts out is the podcast he wants to put out. The corporate pitches he signs off on are ones he would have written. Nothing goes out that he doesn't recognise.

The test. If Ron is asked to review a thumbnail three weeks in, that's a failure. If Ron is asked to approve a nurture email three years in, that's the architecture working.

Scheduled runs store state where Ron's data isn't stored today. [ASK NICK]

Cowork's July 7 release made scheduled tasks and remote sessions real — the reason 04, and the whole rest of the T1 layer, can honestly claim to run without a window open. The catch: those remote runs keep session state in the Claude account, not on the Mac Nick works from. The service-bureau posture — the ongoing fee buys adaptation on Nick's hardware — bends at that seam.

The convenience

Recurring triggers Ron never has to think about. "It just arrives" becomes literal.

The seam

Files-first for outputs (Ron's data stays Ron's files), but the scheduler's own state lives in the Claude account. Options: keep scheduled runs to non-sensitive triggers, or run recurring jobs from Nick's always-on Mac. Nick's call, flagged here so it isn't silently made.

Every automation reads and writes files. That's the whole thing.

No database. No black box. Markdown, CSV, JSON — the shapes any human can open. Amy can audit any decision by reading a file. Sophie can inspect any pipeline by opening the folder. Ron's data stays Ron's files, always.

Why this matters more than it sounds like it does. It's the exit guarantee. If this engagement ends tomorrow, Ron keeps everything the machine built — because everything the machine built is already sitting in his Drive, in the same folder structure his workflow doc points at. The IP is inspectable, the adaptation is the service. No lock-in, no ransom, nothing that behaves like a black box.

Every pipeline step that touches text writes its output to a file first, then a view renders it. The dashboard is optional; the files aren't.

Where the machine is guessing — flagged, not silently resolved.

Every flag below is an invitation to say "no, actually." That's how the engagement calibrates. Ron's edits are the calibration data; his silences are calibration data too, but a spoken correction is a shortcut.

[ASK RON]

The canonical category count. The room has been saying 21 or 27 depending where you look. Ron used 27 on the discovery call; the catalogue supports both 26 (split) and 27 (with one unpopulated). The flag stays open here because the answer belongs to Ron, not to us.

[ASK RON]

Where the Holistic Alternative Broadcasters conversation should sit — in or out of this room. Not discussed on the call. Open by design.

[ASK RON]

Fluoridation as a stance — in the house-style rulebook, out of it, or neither. Same answer: Ron's, not ours.

[ASK RON]

The path-to-yes on new engagements — a standard we could publish, or the shape Ron keeps in his head. Both are legitimate answers; the flag stays open until Ron picks.

[ASK NICK]

Where scheduled-run state lives. Named in the second tension above. Nick's call before any scheduled task carrying sensitive triggers ships.

[ASK SOPHIE]

The GHL migration timing. Three T1-conditional automations (05, 07, 11) wait on it. What order does she want to build in, and where does this room's plumbing help vs. get in the way?

"The interventions are remarkably simple, sustainable, achievable and effective. And effective, that's really important."— Ron, on the discovery call. The design principle applied to his work, applied here to his week.