Playwright disables normal background throttling, so a hidden 5chan page keeps doing P2P and rendering work after a check finishes. Agents verifying in parallel across worktrees stacked whole browser engines on one machine. Add scripts/pw-session.sh, a wrapper that permits one active Playwright browser at a time and records who holds it: - The lock is machine-wide, not per-repository, because the contended resource is RAM and CPU. Every worktree and checkout shares one slot. - Acquisition is an atomic mkdir. Stale locks clear themselves: `open` reclaims any slot whose recorded browser is no longer `status: open` in `playwright-cli list --all`, so an interrupted workflow cannot strand the budget. When that list cannot be read the lock is left alone, so a broken CLI never silently disables the budget. - `open` exits 75 when the slot is busy; `--wait[=SECONDS]` blocks instead. - `close` always stops the browser, even when the lock was already lost, and never releases a slot held by a different session. - `status` reports the holder and whether its browser is still alive. Agent policy now runs browser engines and profiler batches sequentially, uses Chrome/Blink during iteration and the full engine matrix only for final verification, and never uses `close-all` or `kill-all` while other agents may own sessions. Covered by scripts/pw-session.test.js.
4.2 KiB
name, description
| name | description |
|---|---|
| inspect-elements | Resolve on-screen 5chan DOM elements to React source files, line numbers, component names, and ownership stacks using the app's dev-only element-source helpers and playwright-cli. Use when an agent needs to inspect a page element, map a snapshot ref to source code, confirm which component rendered a node, or follow up after $profile-browsing finds a rerender hotspot and needs file-level attribution. |
Inspect Elements
Use this skill to jump from a concrete DOM node in the running 5chan app to the React file and component stack that produced it.
Prerequisites
- Dev server running at
https://5chan.localhost playwright-cliinstalled- Use the local dev app, not production. The element-source helpers are only exposed in dev mode.
Quick workflow
- Open the target route with
./scripts/pw-session.shso the shared browser slot is respected. - Run
playwright-cli snapshotand choose the relevant element ref. - Resolve that ref through the app helper:
playwright-cli -s=inspect eval "async el => JSON.stringify(await window.__ELEMENT_SOURCE__.resolve(el))" e7
The result includes:
source: the most useful file/line match for the elementcomponentName: the nearest meaningful React componentstack: ownership stack from the concrete node upwardtagName: the underlying DOM tag
Session setup
./scripts/pw-session.sh open inspect https://5chan.localhost
playwright-cli -s=inspect goto https://5chan.localhost/all
playwright-cli -s=inspect eval "window.__ELEMENT_SOURCE__?.ready ?? false"
playwright-cli -s=inspect snapshot
If ready is false, wait a moment and evaluate again. If window.__ELEMENT_SOURCE__?.error is set, report that error instead of continuing.
Resolve strategies
Prefer snapshot refs because they target the exact live DOM node you just inspected.
Snapshot ref
playwright-cli -s=inspect eval "async el => JSON.stringify(await window.__ELEMENT_SOURCE__.resolve(el))" e7
Selector
Use this only when the element is easy to target and a snapshot ref is not practical.
playwright-cli -s=inspect eval "JSON.stringify(await window.__ELEMENT_SOURCE__.resolveBySelector('[data-testid=\"composer\"]'))"
Screen coordinates
Useful when you have a screenshot or a visually obvious hotspot.
playwright-cli -s=inspect eval "JSON.stringify(await window.__ELEMENT_SOURCE__.resolveAtPoint(320, 420))"
Format the ownership stack
playwright-cli -s=inspect eval "async el => { const info = await window.__ELEMENT_SOURCE__.resolve(el); return JSON.stringify({ ...info, formattedStack: window.__ELEMENT_SOURCE__.formatStack(info.stack, 5) }); }" e7
Use formattedStack when you need a short, readable trace for the final report.
Close the session immediately after collecting the needed source evidence, including when resolution fails:
./scripts/pw-session.sh close inspect
Profiling follow-up
When $profile-browsing reports a hot route or rerender-heavy area:
- Reopen the route in a fresh Playwright session through
./scripts/pw-session.sh. - Snapshot the concrete list item, card, modal, or toolbar node that looks relevant.
- Resolve it with
window.__ELEMENT_SOURCE__.resolve(...). - Use
source.filePathas the direct edit target andstackto understand parent ownership.
This is a complement to react-scan, not a replacement. react-scan tells you which components rerender too often. inspect-elements tells you which exact source file produced the node you are looking at.
Rules
- Prefer snapshot refs over brittle selectors.
- Inspect the actual node the user cares about, not a distant wrapper, unless wrappers are the suspected problem.
- If
sourceis null butstackexists, use the first useful stack frame rather than guessing. - If both
sourceandstackare empty, report that the node could not be resolved and pick a nearby parent element instead. - If the browser slot is held, retry after the owning workflow finishes or block on
./scripts/pw-session.sh open --wait ...; do not bypass the lock or useclose-all/kill-all. - Close the exact named session in a finally-style cleanup.