Dispatch / Publishing schedule

We publish the work after the evidence.

Each week, we choose a change worth explaining. We show what happened, what we learned, and where the result has limits.

  • Source Repository
  • Lesson One per week
  • Claim Evidence-gated

Working rhythm

One lesson, carried across the right formats.

We share updates on LinkedIn, demonstrate a full workflow on YouTube twice a month, and send selected lessons by email.

Tuesday

LinkedIn

One practical lesson from verified work.

Thursday

LinkedIn

The artifact, receipt, or boundary behind the lesson.

Alternate Thursdays

YouTube

A complete 5–8 minute workflow walkthrough.

Friday

X / Instagram

A technical adaptation or an existing demonstration clip.

Up to twice monthly

Email

A validated field note in the existing CREATE SOMETHING archive.

Current editorial queue

Four questions from work already in the repository.

These dates are plans. We publish only after reviewing the supporting evidence and checking what can be shared.

01

Week of September 7, 2026

Can an operator see what an agent changed?

Inspectable canvas changes. Source: Draw semantic composition and scoped production tools.

Evidence
Public-safe canvas walkthrough with one bounded change
Channels
LinkedIn · YouTube · X · Instagram
State
evidence review
02

Week of September 14, 2026

How do you check a code claim before repeating it?

Verification before claims. Source: Ground calibration and release verification.

Evidence
One reproducible check with the exact result and limit
Channels
LinkedIn · X
State
evidence review
03

Week of September 21, 2026

Does done mean prepared, approved, executed, or verified?

A receipt must name the state. Source: Workflow receipt history and the existing field note.

Evidence
Synthetic workflow from request through wait, stop, and receipt
Channels
LinkedIn · YouTube · Email · X
State
evidence review
04

Week of September 28, 2026

What should remain when the agent leaves?

The next operator needs a usable handoff. Source: Atlas/Substrate operating loop and Agent Foundation handoff.

Evidence
Public-safe map with owner, boundary, proof, and recovery
Channels
LinkedIn · X · Instagram
State
evidence review

Publication boundary

A commit is the start of the story.

01

Git establishes what changed; a merge does not establish live behavior.

02

The agent wiki helps locate the operating lesson; its source artifacts remain authoritative.

03

CTX can explain earlier reasoning; current source, tests, and live behavior gate the claim.

04

Every public example names what the evidence proves and what remains unknown.

05

Client-private records and access changes do not become public stories.

Bring the next workflow

Show us the handoff your team still rebuilds by hand.

We’ll look at the task, tools, approvals, and expected result before proposing a build.

Owner
Workflow owner
Authority
Human approval
Proof
Map + evidence plan
State
ready
AI workflow system

Find the task where AI could help.

Start with one repeated task. Agree on what AI can do, who checks its work, and how to measure the result.

Verified field result

Automation prepared the evidence. Human judgment still decided.

The system gathered evidence for reviewers. A person still decided whether to approve or reject each submission.

49/ 50
49 of 50 selected cases produced usable evidence packets for human decision.
Workflow
Marketplace template review
Receipt
#FR-2026-01
Verified
May–June 2026
Changes to external systems
0
  1. Map the frictionList the tools, steps, and person responsible.
  2. Test one pathTest the routine work and check that approvals still work.
  3. Keep the proofReview what ran, what stopped, and what needs recovery.