Polish Put a working feature in front of the user and turn their live observations into focused UX fixes on the running page. Two ways to collect those observations: traditional , where the user types what could be better; and live , where they talk and draw on the page while a voice interviewer turns speech into units and you act on them at checkpoints. Done: the user ends the polish loop, every requested fix is reflected in the live feature or reported as blocked, and the in-scope changes are saved in local commit(s). A live session also ends with a residual list of the units that were not…

; then\n echo \"$local_port\"\n exit 0\n fi\n fi\n done\nfi\n\n# ── Probe 3: Rails config/puma.rb ───────────────────────────────────────────\n\nif should_probe \"$PROJ_TYPE\" \"puma\"; then\n puma_file=\"$PROJECT_ROOT/config/puma.rb\"\n if [ -f \"$puma_file\" ]; then\n puma_port=$(grep -Eo 'port[[:space:]]+[0-9]+' \"$puma_file\" 2>/dev/null | head -1 | grep -Eo '[0-9]+')\n if [ -n \"$puma_port\" ]; then\n echo \"$puma_port\"\n exit 0\n fi\n fi\nfi\n\n# ── Probe 4: Procfile.dev ───────────────────────────────────────────────────\n\nif should_probe \"$PROJ_TYPE\" \"procfile\"; then\n procfile=\"$PROJECT_ROOT/Procfile.dev\"\n if [ -f \"$procfile\" ]; then\n # Extract the web line\n web_line=$(grep -E '^web:' \"$procfile\" 2>/dev/null | head -1)\n if [ -n \"$web_line\" ]; then\n # Match -p \u003cn>, -p\u003cn>, --port \u003cn>, -p=\u003cn>, --port=\u003cn>\n proc_port=$(printf '%s' \"$web_line\" | grep -Eo '(-p[= ]*|--port[= ]+)[0-9]+' | head -1 | grep -Eo '[0-9]+')\n if [ -n \"$proc_port\" ]; then\n echo \"$proc_port\"\n exit 0\n fi\n fi\n fi\nfi\n\n# ── Probe 5: docker-compose.yml ─────────────────────────────────────────────\n\nif should_probe \"$PROJ_TYPE\" \"docker-compose\"; then\n compose_file=\"$PROJECT_ROOT/docker-compose.yml\"\n if [ -f \"$compose_file\" ]; then\n # Simple line-anchored grep for port mappings: - \"NNNN:NNNN\" or - NNNN:NNNN\n compose_port=$(grep -Eo '\"[0-9]+:[0-9]+\"' \"$compose_file\" 2>/dev/null | head -1 | grep -Eo '[0-9]+' | head -1)\n if [ -n \"$compose_port\" ]; then\n echo \"$compose_port\"\n exit 0\n fi\n fi\nfi\n\n# ── Probe 6: package.json scripts ───────────────────────────────────────────\n\nif should_probe \"$PROJ_TYPE\" \"package-json\"; then\n pkg_file=\"$PROJECT_ROOT/package.json\"\n if [ -f \"$pkg_file\" ]; then\n # Look for --port or -p in dev/start scripts\n pkg_port=$(grep -Eo '(-p[= ]+|--port[= ]+)[0-9]+' \"$pkg_file\" 2>/dev/null | head -1 | grep -Eo '[0-9]+')\n if [ -n \"$pkg_port\" ]; then\n echo \"$pkg_port\"\n exit 0\n fi\n fi\nfi\n\n# ── Probe 7: .env files ─────────────────────────────────────────────────────\n\nif should_probe \"$PROJ_TYPE\" \"env\"; then\n for envfile in \\\n \"$PROJECT_ROOT/.env.local\" \\\n \"$PROJECT_ROOT/.env.development\" \\\n \"$PROJECT_ROOT/.env\" \\\n ; do\n env_port=$(parse_env_port \"$envfile\")\n if [ -n \"$env_port\" ]; then\n echo \"$env_port\"\n exit 0\n fi\n done\nfi\n\n# ── Probe 8: Framework default lookup table ──────────────────────────────────\n\nif should_probe \"$PROJ_TYPE\" \"default\"; then\n case \"$PROJ_TYPE\" in\n rails|next|nuxt|remix|procfile|\"\")\n echo \"3000\"\n ;;\n vite|sveltekit)\n echo \"5173\"\n ;;\n astro)\n echo \"4321\"\n ;;\n *)\n echo \"3000\"\n ;;\n esac\n exit 0\nfi\n\n# Final fallback (should not normally be reached)\necho \"3000\"\nexit 0\n","content_type":"application/x-sh; charset=utf-8","language":"bash","size":10211,"content_sha256":"176895fd7fded26fb1ce1394d07a0ca6308a2fa155b7f230c7646f94abc0b4b5"}],"content_json":{"type":"doc","content":[{"type":"heading","attrs":{"level":1},"content":[{"text":"Polish","type":"text"}]},{"type":"paragraph","content":[{"text":"Put a working feature in front of the user and turn their live observations into focused UX fixes on the running page. Two ways to collect those observations: ","type":"text"},{"text":"traditional","type":"text","marks":[{"type":"strong"}]},{"text":", where the user types what could be better; and ","type":"text"},{"text":"live","type":"text","marks":[{"type":"strong"}]},{"text":", where they talk and draw on the page while a voice interviewer turns speech into units and you act on them at checkpoints.","type":"text"}]},{"type":"paragraph","content":[{"text":"Done:","type":"text","marks":[{"type":"strong"}]},{"text":" the user ends the polish loop, every requested fix is reflected in the live feature or reported as blocked, and the in-scope changes are saved in local commit(s). A live session also ends with a residual list of the units that were not applied and the path of its session log. A server or checkout blocker also ends the run when it is reported with the evidence needed to resume.","type":"text"}]},{"type":"paragraph","content":[{"text":"Boundaries:","type":"text","marks":[{"type":"strong"}]},{"text":" the user drives what to inspect and change; do not invent an autonomous checklist or expand into general QA. Never work on the repository's default branch. This workflow may edit and locally commit the requested polish, and in live mode the disclosed riffrec setup commit, but it never pushes or opens a PR.","type":"text"}]},{"type":"heading","attrs":{"level":2},"content":[{"text":"Run","type":"text"}]},{"type":"ordered_list","attrs":{"order":1,"listStyle":"number"},"content":[{"type":"list_item","content":[{"type":"paragraph","content":[{"text":"Ask live or traditional.","type":"text","marks":[{"type":"strong"}]},{"text":" Ask once, before any server starts. Include this disclosure with the question: live mode needs an OpenAI key in the environment, runs a voice interviewer in the page that hears the user, is told what they click and draw on, and looks at their screen when they point at something or ask it to, streams the session to a local endpoint this skill runs, and, when riffrec is not yet in the app, adds the riffrec dependency and a provider mount as a setup commit on the current branch that stays after the session. Traditional continues with steps 2–5 unchanged. Live: read ","type":"text"},{"text":"references/live-start.md","type":"text","marks":[{"type":"code_inline"}]},{"text":"; it owns preconditions, install, the endpoint, and the URL handoff, applies the workspace and server rules of ","type":"text"},{"text":"references/run.md","type":"text","marks":[{"type":"code_inline"}]},{"text":" itself, then routes to ","type":"text"},{"text":"references/live-loop.md","type":"text","marks":[{"type":"code_inline"}]},{"text":" for the session and its close. Steps 2–5 do not run for a live session.","type":"text"}]}]},{"type":"list_item","content":[{"type":"paragraph","content":[{"text":"Get the live page ready.","type":"text","marks":[{"type":"strong"}]},{"text":" Read ","type":"text"},{"text":"references/run.md","type":"text","marks":[{"type":"code_inline"}]},{"text":" before resolving the requested ref or starting anything. It owns existing-worktree safety, dev-server discovery, the bundled-script calls, reachability, and the browser handoff.","type":"text"}]}]},{"type":"list_item","content":[{"type":"paragraph","content":[{"text":"Wait for observations.","type":"text","marks":[{"type":"strong"}]},{"text":" Tell the user where the server is running and ask what could be better. Do not start a review pass while they browse.","type":"text"}]}]},{"type":"list_item","content":[{"type":"paragraph","content":[{"text":"Iterate.","type":"text","marks":[{"type":"strong"}]},{"text":" For each requested change, inspect only as needed, edit the in-scope surface, and let hot reload update the page. When the user asks you to inspect the result, use a browser capability available in the active harness; if none exists, ask them to describe what they see.","type":"text"}]}]},{"type":"list_item","content":[{"type":"paragraph","content":[{"text":"Close locally.","type":"text","marks":[{"type":"strong"}]},{"text":" When the user says they are done, invoke ","type":"text"},{"text":"ce-commit","type":"text","marks":[{"type":"code_inline"}]},{"text":" for the polish changes, then report the commit(s), the still-running server URL, and any residual blocker.","type":"text"}]}]}]}]},"metadata":{"date":"2026-11-02","name":"ce-polish","author":"@skillopedia","source":{"stars":25863,"repo_name":"compound-engineering-plugin","origin_url":"https://github.com/EveryInc/compound-engineering-plugin/blob/HEAD/skills/ce-polish/SKILL.md","repo_owner":"EveryInc","body_sha256":"985888097d7e82a04a4b59856d53d7e6a30854094400b5be36c278261f0c92fa","cluster_key":"c7c982cbc2d06e9f73917d605274e32375a6c64fe1b98835a2e396b58307b50a","clean_bundle":{"format":"clean-skill-bundle-v1","source":"EveryInc/compound-engineering-plugin/skills/ce-polish/SKILL.md","text_bytes":259439,"source_commit":"77cc431daaebea7ee9f33ac530a66eec5404572b","skills_sh_rank":12149,"attachment_count":23,"text_attachments":23,"attachment_storage":"skill_attachments","binary_attachments":0},"cluster_size":1,"skill_md_path":"skills/ce-polish/SKILL.md","import_metadata":{"date":"2026-11-02","author":"@skillopedia","version":"v1","category":"design-ux","category_label":"Design"},"exact_dupes_collapsed_into_this":0},"version":"v1","category":"design-ux","import_tag":"skills-sh-top20k-2026-11","description":"Polish a working feature through user-directed live browser feedback. Use when a functional feature needs focused UX refinement before shipping.","argument-hint":"[PR number, branch name, or blank for current branch]","disable-model-invocation":true}},"renderedAt":1793661602139}

Polish Put a working feature in front of the user and turn their live observations into focused UX fixes on the running page. Two ways to collect those observations: traditional , where the user types what could be better; and live , where they talk and draw on the page while a voice interviewer turns speech into units and you act on them at checkpoints. Done: the user ends the polish loop, every requested fix is reflected in the live feature or reported as blocked, and the in-scope changes are saved in local commit(s). A live session also ends with a residual list of the units that were not…