Briar, we need to get some work done
I want Briar to help us get good work done. We found that some of the instructions meant to help were keeping them busy with the wrong things.
Read and shareChop Wood Carry Water
Durable Agent Harness
I’ve been building a working relationship with coding agents — now Codex-first, with portable skills, memory, signed patterns, human gates, and proof close to the work. Models and cockpits change; the practice should survive them. This site is the public notebook.
Agents: start here — pack, Priority 0 (P0) rules, and the 60-minute path.
Where would you like to start?
Try one checked result, see how the setup works, or read what happened on real engineering jobs.
Try itRun a failing test, an attempted correction and a verified fix. A small exercise with source.
Understand the setupChoose Codex or Cursor, check the boundaries and do one useful task.
Read the engineering storiesActual work, mistakes, costs and what changed.
The reference sections remain below. Tell us about a step that needs improving.
Start in 60 minutes
Guidance reviewed ; historical snapshot dates below are unchanged.
Try the runnable exercise · Verify downloads and browse file manifests · Feedback on a step
Then pick another door
You don’t have to read the whole notebook in one go.
Do
Playbook
Fifteen steps if I were starting from scratch tomorrow.
Move
What Codex changed
Portable skills, layered guidance, worktrees, self-evals — and the parts that did not transfer.
Copy
Starter kit
Thin templates — skills, rules, patterns, lanes, clocks.
Learn
Kaizen → Toyota Way Practices
Improve on purpose; name the practice that matches the pain.
Run
Preloop + OpenRouter
Runtime policy under human gates; route models when it helps.
Ops
Channels & lean
Messaging, ingest, token rent, practice surfaces.
Browse
Explore
Timeline, stack, capability, patterns, glossary, research.
Track
Measure
Dated snapshots and examples, with their limits.
Read
Insights
What broke, what locked, what earned a place on disk.
The cockpit changed. The work did not.
What the interim earned
What did not transfer automatically
Build your own working relationship
Anti-patterns
Kaizen → Hansei → Toyota Way Practices
MemPalace for memories
Do
Not
Toyota Way Practices — member practices
Runtime layers beyond the cockpit
PR / commit checks
Do
Not
Channels, media & token lean
Messaging, media ingest, token rent, and practice surfaces — one operating system, same human gates.
Do
Not
Do
Not
Anti-patterns
Explore
One panel at a time — timeline, stack, capability, patterns, glossary, research.
Case studies
Measurement kit
Four clocks
Humbling baseline: METR’s early-2025 RCT found experienced open-source developers were about 19% slower on familiar issues when AI tools were allowed — while still believing they were faster. That does not kill the case for agents; it kills lazy multipliers.
Comparison bands
When a review is useful
Starter kit
Insights
How I work with Briar, the skills harness I’m creating
An introduction for embedded engineers: how we work with Briar, what the harness does, what has worked, where it gets difficult and how to try it.
Read and shareFinding the Sweet Spot
You ask your assistant to take something off your list. Somehow, keeping them on track becomes another job.
Read and shareRemember loving Napster?
I shared an article with Briar on Telegram and asked for the music in Spotify. About twelve minutes later, I had 15 albums ready to play.
Read and shareThe Token Bonfire - when things get out of control
Three free ChatGPT resets and another £75 gone within hours. How our coding agents complicated our work, and what we changed in Briar.
Read and shareThe Model Is Only Part of the Machine
Follow a Yocto build failure through an LLM harness and its tools: what runs, what enters context, and what we still need to check on the board.
Read and shareAbout
I’m Alex Lennon (@embedded_iot). This notebook is engineering practice and insights in public — clocks, playbook, patterns, cases and starters. Steal what helps; measure your own results. AI-assisted; use at your own risk.
Public baseline v1.2.0 (2026-09-12) — token-disciplined, Codex-first, Cursor-compatible — baseline · v1.1 history. Updates land in the end-of-week review when the week earned them.
Content CC BY-SA 4.0 · code MIT · LICENSE