Takeaway: Next time I'd give the full spec plus real breakpoint screenshots up front instead of discovering overflow and mobile issues one screenshot at a time — that back-and-forth on the projects grid alone took five different layout attempts before landing on the paginated 3-column version.
Takeaway: Next time I'd open with the fact that Reddit blocks server-side requests entirely and only trusts real browser cookies — that alone would have saved the whole detour through proxy routes, Windows path bugs, and Playwright dead-ends.
Takeaway: Next time I'd paste the actual CLAUDE.MD requirements (RemixJS) into the very first prompt instead of assuming the agent read it correctly — that one missed instruction cost the most time, especially once it started patching node_modules to fight Windows file locks.
Takeaway: Next time I'd tell it upfront to just use agy for background workers since pi -p silently hangs without doing real work — that cost a good chunk of the session polling dead processes.
Takeaway: I'd start by asking it to grep the installed package for HEADROOM_ env vars directly instead of trusting the hosted docs, since those turned out to be too vague to use.
Takeaway: I'd start the next audit by explicitly telling it to re-verify the NEVER DO list against reality before touching anything, since two of the entries there turned out to be stale (pacnew notes, BIOS assumption).
Takeaway: Next time I'd tell it upfront to use a single server-side fetch helper (or a proxy +server.ts route) instead of env vars for backend URLs, since that's what ate an hour of back-and-forth.
Takeaway: Next time I'd nail down the SvelteKit server-fetch/env-var pattern up front instead of letting the agent iterate through three different fixes for the same root cause.
Takeaway: Next time I'd front-load the constraint that Groovy specs compile lambdas to opaque proxies, since that cost a whole probing detour before the nested-selector feature could be built correctly.
Takeaway: Next time I'd open with the hardware constraints up front — the USB-OTG direct-connect dead end cost real time before we found the powered-hub workaround, and I should've mentioned the DSP needs a hub from the start.
Takeaway: Next time I'd ask upfront how this OS's other interactive JS apps handle input, since the busy-poll getKeyNonBlock() loop turned out to be the real cause of the exit crash, not the VGA I/O port theory the agent chased first.
Takeaway: Next time I'd mention upfront that the shared layout hits the backend for /counter and /models, so previewing any page needs either the full stack or a stub — that cost real time here.
Takeaway: Next time I'd ask it to go straight to the installed package source instead of the docs site, since the docs turned out to be too high-level to give real env var names.
Takeaway: Next time I'd tell it upfront to sanity-check that pi workers are actually producing output within a minute or two, since three of them silently hung for over five minutes doing nothing while I waited.
Takeaway: Next time I'd ask upfront whether the messaging platform requires a reply anchor for topic delivery before wiring cron jobs to a named topic — that assumption cost two failed delivery attempts.
Takeaway: Next time I'd put the taste-notes requirement up front instead of bolting it on after a full render cycle — the agent did solid disciplined work delegating to sub-workers and gating changes, but the real decisions were always going to need my eyes on the video first.
Takeaway: Next time I'd ask it to run the TypeScript check earlier and more often instead of writing all the frontend files first and fixing a pile of errors at the end.
Takeaway: Next time I'd point straight to the migrate_votes.py file and backend/config.py POSITIVE_PREFS/NEGATIVE_PREFS since that's what actually answered the question.
Takeaway: Next time I'd spell out the full spec up front — 'fetch Shakespeare books sized to 1.2x the corpus' — instead of letting it start running on a vague version first.
Takeaway: Next time I'd start with 'trace the actual config precedence and provider fallback logic, don't just read the top-level JSON' — that's what caught the real bug (searchProvider silently overriding provider).
Takeaway: Next time I'd tell it up front to match the existing convention in develop before inventing new schema entries like licenses, since checking real precedent is what actually fixed things.
Takeaway: Next time I'd start the prompt by naming the exact Reddit endpoints and noting upfront that reddit.com blocks server-side fetches without browser cookies — that alone would have saved the long detour through proxies, User-Agent spoofing, and CORS services.
Takeaway: Next time I'd tell it up front to match the existing input-loop pattern (blocking getChar, like quill and the shell) instead of letting it invent a getKeyNonBlock busy-poll — that's what caused the GPF on quit.
Takeaway: Next time I'd start by dumping the raw config files myself instead of pasting agent transcripts and letting Claude re-diagnose from scratch — it wasted a lot of turns re-reading settings.json across live and repo copies.
Takeaway: Next time I'd point it straight at the extensions.md/themes.md docs and the smoke-test approach from the start, since digging through docs and installed packages ate most of the session.
Takeaway: Next time I'd bake in a retry/resume path for the workers up front, since two of three codex workers died mid-task on SSE timeouts today — the thing that actually saved the round was that I'd told them to leave artifacts on disk, so I could salvage the partial forensics and grade data myself instead of restarting from zero.
Takeaway: Next time I'd just tell it upfront that the site is a static root-level app with no bundler dependency, so it doesn't waste time inspecting main.js and package.json to figure that out.
Takeaway: Next time I'd specify upfront that I want both UEFI/MB2 and legacy BIOS/MB1 to keep working, since that dual-path requirement is what caused most of the debugging churn around the framebuffer, paging, and GRUB modules.
Takeaway: Next time I'd ask for the stack-size and static-storage constraints to be checked up front, since two of the three crashes came from oversized structs living on the stack rather than logic bugs.
Takeaway: Next time I'd ask upfront for the stack-usage rules (no post-init malloc, bounded everything) to be double-checked against test harness patterns, since two of the three real bugs were stack overflows from large structs being copied or placed on the call stack instead of static storage.
Takeaway: Next time I'd nail down the QEMU/OVMF/WSL boot pipeline first before layering framebuffer and GRUB experiments on top of it — half the pain was Windows/WSL path translation and stale OVMF vars, not the actual kernel or userland code.
Takeaway: I'd paste the exact error PowerShell throws next time instead of just saying 'fix this' so the agent doesn't have to guess which line is broken.
Takeaway: Next time I'd tell it upfront to match the existing sibling model entries' conventions (license, fyi text) instead of inventing new fields, since that's what caused the rework.
Takeaway: Next time I'd put the milestone list, the GitHub-vs-Codeberg cryptoscam split, and the Python-version target (3.7+) all in the first prompt instead of drip-feeding them — the agent had to backtrack on dataclasses and kw_only fields because 3.7 support came in late.
Takeaway: Next time I'd give the exact copy (song, project URLs, email, usernames) up front instead of patching them in one by one at the end — that ate a lot of turns.
Takeaway: Next time I'd tell it upfront to always launch background threads (HTTP listener, per-connection handlers) as self-cleaning/detached instead of calling thread.destroy, since that single join call silently blocked the whole game window from ever opening.
Takeaway: Next time I'd specify target screen sizes and content lengths up front, since the ABOUT overflow only showed up once I actually looked at the rendered page.
Takeaway: Next time I'd put the full bokeh-fix-plus-breathing-room task in one prompt from the start instead of splitting it into scout/backup/implement stages — the orchestrator ended up doing that anyway.
Takeaway: Next time I'd flag up front that any CLI flags removed during a refactor need a compatibility shim, since removing --recover silently broke one test fixture.
Takeaway: Next time I'd specify upfront that I want the split and training pipeline to preserve the conversation metadata (system/human/gpt) rather than letting the agent default to the flattened corpus.txt — that correction is what forced the whole pipeline to be rebuilt properly.
Takeaway: Next time I'd tell it upfront that Stage 3's 'parser/bytecode boundary' can't be a real physical separation in a 63K-line monolith like quickjs-ng, so it skips straight to building the JS_WriteObject/JS_ReadObject wrapper instead of discovering that mid-session.
Takeaway: I'd include the skills note in my first prompt since I had to add it as a follow-up anyway.
Takeaway: Next time I'd specify upfront that a .env.example should be verified with a read-back before relying on it, since the 'file created' message wasn't trustworthy on its own.
Takeaway: Next time I'd just ask it to check whether the project even needs a build step before touching vercel.json — that's the fix that actually mattered, not the plugin install.
Takeaway: Next time I'd spell out exact screen resolutions and the grid/sidebar split ratio in the very first prompt instead of iterating three times on layout — that back-and-forth cost more turns than any of the actual milestone work.
Takeaway: Next time I'd tell it up front that font files already live locally in fonts/ so it doesn't default to a CDN import.
Takeaway: I'd paste the exact error message next time so the agent doesn't have to guess which line was actually broken.
Takeaway: Next time I'd put the cover-photo disposition check and the taste-notes review earlier in the same prompt, since the disposition bug (attached_pic=0) only got caught because I forced an ffprobe verification step right after muxing — that's the correction that actually mattered here.
Takeaway: Next time I'd tell it upfront that melt and xmllint passing doesn't mean kdenlive itself will open the file cleanly — actually launching the real app was the check that mattered, not the automated gate.
Takeaway: Next time I'd put the actual framework name in the first prompt instead of relying on the agent to read CLAUDE.MD correctly - that Remix mixup cost a lot of time patching node_modules on Windows.