Promptprint
Promptprint · 2026·02 — 2026·06
700

From one-line asks to multi-step system design

Over five months (Feb–Jun 2026) you asked 700 questions across Claude Code and Codex. It started with short "how do I…" lookups — roughly 34 characters each — and by late spring the average question grew to ~272 characters: multi-step specs with code blocks and explicit goals. The vocabulary moved with it, from setup basics (docker, git, pip) to system concerns (latency, cache, throughput). And the way you work with the AI matured — 300 trade-off/critique prompts and 270 verification checks: you started directing and pressure-testing answers, not just dictating tasks. (This is a sample report built from synthetic data.)

claude 490 codex 210Period 5 months
Next Bearings

Where the
Compass Points

Reading the current patterns to find the next focus.

01
Bring code context in earlier

Code blocks show up in ~39% of questions, but almost all of them are in the later months — the early asks were context-light.

Paste the error and the relevant file at the start of a question to cut the back-and-forth before the AI has enough to work with.

02
Pressure-test more, not just verify

Verification asks reached 270, but counter-questions ('why not…', 'rationale') stayed low at 90 — you check answers more than you challenge them.

After a proposal, ask 'why not the simpler option?' to force the AI to defend the design before you commit to it.

Action — Automation

Skills to Build

Repeated, re-explained tasks — build them once as a skill and stop re-asking.

01
tradeoff-review

You ask for the same trade-off comparison over and over — re-describing the options, the goals, and the constraints from scratch each time instead of reusing a fixed frame.

  • 'trade-off' framing recurs ~300 times at an average of 255 chars/question, across 3 months
  • almost all of it is recent — the pattern is still active, not a phase you graduated from
Est. savings Estimate: ~255 chars of re-explanation removed per ask. Assumption — one /skill call replaces the hand-written comparison frame each time; the measured counts (300×, 255 chars) are real, the saving is an estimate.
Skill seed · skill-creator
/skill-creator: build an 'architecture trade-off review' skill. Input: two candidate options + the goal axes (latency / cost / ops load). Output: a side-by-side table on latency, cost, operational load and risk, then a recommendation and the verification questions to run. Representative asks: "kubernetes vs a plain vm for the billing service — what are the trade-offs?", "why not just scale the workers horizontally?".
02
latency-triage

Latency and throughput investigations repeat with the same shape — current numbers, a budget, then 'walk me through the options' — and you rebuild that context every time.

  • 'latency' recurs ~360 times at ~272 chars/question, concentrated in the two most recent months
  • usually paired with a verify/critique ask (check this against the target)
Est. savings Estimate: the repeated p99/budget setup (~270 chars) collapses into one skill call. Assumption — the recurring framing is the reusable part; per-incident numbers still vary.
Skill seed · skill-creator
/skill-creator: build a 'latency triage' skill. Input: current p50/p99 + target + a one-line topology. Output: where the time likely goes, two architectures with trade-offs, and a check against the availability target. Representative ask: "p99 ~800ms, goal under 200ms, keep the cache warm after deploys — walk me through the trade-offs and verify the plan against the goals."
Activity

Chart of Passage

80
2026.02
avg 33 chars
110
2026.03
avg 33 chars
150
2026.04
avg 63 chars
175
2026.05
avg 272 chars
185
2026.06
avg 272 chars

Bar height = monthly question count · number = count for that month

avg questions / session 11.29 · one-shot (single-question sessions) 2% · sessions 62

Question genres — Build 31% · Debug 30% · Understand 17% · Other 17% · Improve 4%

Voyage Log

Log Entries

The phases where your questions shifted.

2026-02 ~ 2026-03

01

Getting set up

Short, practical questions — installing tools, fixing startup errors, basic git. Average ~34 characters. You were learning the terrain.

2026-04

02

Weighing the options

Questions doubled in length and turned comparative: kubernetes vs a VM, redis vs memcached, and the first 'why is this slower?'. You started asking for trade-offs.

2026-05 ~ 2026-06

03

Designing for scale

Multi-step specs with code blocks, latency budgets, and 99.9% availability targets. You set the goals, asked for trade-offs, and made the AI verify its own plan.

Six Dimensions of Growth

Six Bearings

Cross-sections of your working style revealed by your questions.

Topic
Topic Evolution

Your questions migrated from 'how do I set this up' to 'how should this system behave under load'. The early months are full of tooling basics; by May the top terms are almost entirely about runtime behavior.

  • Early terms: docker, git, pip, npm — concentrated in 2026-02 ~ 2026-03
  • 'latency' and 'throughput' first appear in 2026-05
  • 'cache' runs 2026-04 → 2026-06 as the connective theme
Depth
Question Depth

The shift from procedure to principle is the clearest signal in your data: questions got ~8× longer as you moved from 'how do I' to 'why is this slower' to full design trade-offs.

  • Average question length: 34 chars (Feb) → 272 chars (May–Jun), about 8×
  • Length mix moved from mostly 'xs' (under 60 chars) to 'm' (200–600 chars)
  • 'why is X slower than Y' style questions emerge in 2026-04
Meta-skill
AI Meta-skill

You went from taking dictation to running the room — asking for trade-offs, demanding verification, and delegating the final call while keeping the reasoning visible.

  • Critique / trade-off prompts: 300
  • Verification asks ('verify', 'are you sure', 'double-check'): 270
  • Delegation ('your call', 'you decide'): 180; counter-questions ('why not', 'rationale'): 90
Craft
Question Craft

Your prompts became self-contained: code blocks for context and numbered steps that hand the AI a plan instead of a fragment — fewer round-trips per intent.

  • Code blocks in 270 of 700 questions (~39%), almost all in the later months
  • 270 multi-step prompts (numbered 1./2./3. plans)
  • Codex prompts carried code slightly more often than Claude (43% vs 37%)
Mastery
Mastery Traces

Some topics you asked about intensely, then stopped — a sign you internalized them and moved on.

  • 'git' and 'pip' questions cluster in 2026-02 ~ 2026-03, then disappear
  • 'docker' carries from setup (Feb) into the CI discussions (Apr), then fades
  • Later topics (latency, throughput, sharding) only appear once the basics are gone
Clusters
Your Clusters

Three seasons of work surface on their own: a setup season, a comparison season, and a scale season.

  • Setup: docker, git, pip, venv, npm
  • Reliability/compare: kubernetes, redis, postgres, ci, cache
  • Scale/design: latency, throughput, architecture, sharding, availability
Postcards

Cards to Take With You

Click "Save as image" on each card to download it as PNG.