Skip to main content

compare-array-bundle-size

Quickly compare the posthog-js array.js bundle size in the current working tree against a git baseline using the repository's esbuild proxy. Use when asked about array.js size, bundle-size percentage changes, whether browser SDK changes increased the bundle, or for a fast size check that should not run the production Rollup build.

Quellinformationen

Repository
posthog/posthog-js
Letzte Quellaktivität
23. Juli 2026 um 18:38
Erkannte Sprache von SKILL.md
Englisch
Sterne
633
Forks
355

Installationsoptionen

Standardmäßig ist der Prompt ausgewählt, der zuerst die Quelle prüft. Sie können zu einem direkten Befehl wechseln oder eine lokale Kopie herunterladen.

Quelldateien prüfen

Lesen Sie SKILL.md und alle von SkillsMP angezeigten Begleitdateien, bevor Sie sich für eine Installation entscheiden.

SKILL.md wird angezeigt

SKILL.md
Quellanweisungen · Schreibgeschützte Vorschau
name
compare-array-bundle-size
description
Quickly compare the posthog-js array.js bundle size in the current working tree against a git baseline using the repository's esbuild proxy. Use when asked about array.js size, bundle-size percentage changes, whether browser SDK changes increased the bundle, or for a fast size check that should not run the production Rollup build.
compatibility
Requires a posthog-js checkout with dependencies installed.
# Compare `array.js` bundle size Use the repository's fast comparison command. Do not run TypeScript compilation, Terser, Rollup, Turbo, or the full build unless the user separately requests production bundle numbers. ## Workflow 1. Use the baseline named by the user. If none is given, let the script default to `origin/main` and then `main`. 2. From the repository root, run: ```bash pnpm bundle-size:array [baseline-ref] ``` 3. Report: - the resolved baseline ref and commit - minified percentage and byte change - gzip percentage and byte change - Brotli percentage and byte change - elapsed time 4. Lead with the percentage most relevant to the user's question. If they did not specify one, lead with minified and include the compressed results below it. ## Interpretation - The command builds both source trees with the same esbuild version, options, installed dependencies, and workspace-source aliases, so the percentage is an apples-to-apples comparison. - The current side includes tracked, uncommitted, and imported untracked source changes from the working tree. - The baseline side is a temporary source snapshot from the resolved git commit; the command does not check out or modify the working tree. - Treat the result as a fast directional comparison. Absolute bytes differ from the production Rollup/Terser output. - A positive percentage is a size increase; a negative percentage is a reduction. ## Validation and troubleshooting - To verify the comparison machinery itself, run `pnpm bundle-size:array HEAD`. With no relevant uncommitted source changes, every row should report `0.00%`. - If the baseline ref cannot be resolved, ask for another local git ref or fetch it only with the user's approval when network access is required. - If esbuild or other dependencies are missing, explain that dependencies must be installed. Do not substitute the full production build.
Auf GitHub ansehen