Agents & automation · 9 min read
A small, inspectable agent exercise: choose a model, define the job, test a fictional incident, and decide whether a second agent earns its place.
Updated 2026-10-16
Knowledge work · 7 min read
Give a persistent agent a clear project, define meaningful updates, and test whether it can maintain a source-linked decision log.
Updated 2026-10-16
Techniques · 11 min read
Choose skills, workers, retries and explicit state using a worked brief-writing task, then test whether your workflow notices a missing check.
Updated 2026-10-16
Learning & research · 8 min read
A small first run, a worked SQL exercise, and an independent check that separates a passing answer from actual understanding.
Updated 2026-10-16
Creative work · 8 min read
A practical scene-planning exercise with an asset sheet, shot brief, controlled revisions, and a continuity review. Discovered through GitHub’s changing project feed.
Updated 2026-10-16
Creative work · 7 min read
What Motion makes, who can access it, and a complete brief-to-MP4 exercise with a storyboard, prompts, and a review checklist.
Updated 2026-10-16
Everyday assistants · 6 min read
A fictional weekly coordination exercise with source IDs, conflicting dates, explicit ownership and a correction test.
Updated 2026-10-16
Agents & operations · 8 min read
Set up a focused Bot, produce a reviewable business result, and learn when a second Bot actually helps.
Updated 2026-10-16
Everyday assistants · 8 min read
Try a life-admin task with a concrete review sheet, then understand what the new Poppy protocol draft does—and what it does not establish about your account.
Updated 2026-10-16
Games & creative workflows · 10 min read
Build a consistent asset family, then use FLUX.3 reference-edit controls without losing silhouettes, pivots or the ability to review changes.
Updated 2026-10-16
Documents & practical evaluation · 8 min read
A fictional receipt exercise with page evidence, a local serving path, arithmetic checks and deliberate failure cases.
Updated 2026-10-16
Everyday assistants · 7 min read
A six-row CSV, exact expected totals, and a repeatable way to check whether a local assistant understands dates, refunds and transfers.
Updated 2026-10-14
Creative work · 9 min read
Direct a two-second wave, keep the pose changes editable, and review the exported frames before putting the animation in a game or video.
Updated 2026-10-14
Audio & creative work · 6 min read
Build a short workshop announcement, compare takes fairly, and keep a usable handoff for the next edit.
Updated 2026-10-13
Knowledge work · 7 min read
Build a small invoice-flow diagram, correct one changed requirement, and hand over both a readable PDF and editable evidence.
Updated 2026-10-13
Everyday assistants · 7 min read
Check the October 12 Muse model availability, then evaluate a personal-agent task with source evidence and a deliberate memory correction.
Updated 2026-10-13
Developer practice · 6 min read
Install the documented Claude Code plugin, then test retrieval against an original incident log with two failures and a misleading success message.
Updated 2026-10-12
Creative practice · 6 min read
A complete prompt-to-test exercise for an original delivery game, with a state table, revision prompts and a playtest sheet.
Updated 2026-10-12
Development & investigation · 8 min read
A scoped REA investigation, a comparison harness, and lessons from a separate long-running game-decompilation report—including how to check the checker.
Updated 2026-10-12
Design & physical prototypes · 9 min read
Build a small desk-label plate from a precise brief, inspect its geometry, change one dimension, and hand off editable source with measured exports.
Updated 2026-10-12
Everyday assistants · 7 min read
Set up a local dictation trial, compare cleanup modes, catch changed names and commitments, and decide whether speaking actually saves you editing time.
Updated 2026-10-11
Creative work · 8 min read
Build a twelve-second release card from editable HTML, inspect its timing, export an MP4, and make one controlled revision without regenerating the whole story.
Updated 2026-10-11
Knowledge work · 7 min read
A six-slide decision brief with fictional data, a claim ledger, precise generation prompts, and an editability test for the exported PowerPoint.
Updated 2026-10-11