SOC 職業分類に基づく
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
直接コマンドでは確認用 Prompt が省略されます。実行前にソースを確認してください。
npx skills add https://github.com/hashintel/brunch --skill d3kコマンドは1行のまま表示されます。コピー前に横へスクロールして全体を確認してください。
ローカルで確認しますか？SkillsMP が現在取得できるファイルをダウンロードできます。
SKILL.md を表示中
| name | d3k |
| description | d3k assistant for debugging web apps |
d3k captures browser and server logs in a unified log file. Use these commands:
d3k errors # Show recent errors (browser + server combined)
d3k errors --context # Show errors + user actions that preceded them
d3k errors -n 20 # Show last 20 errors
d3k logs # Show recent logs (browser + server combined)
d3k logs --type browser # Browser logs only
d3k logs --type server # Server logs only
d3k fix # Deep analysis of application errors
d3k fix --focus build # Focus on build errors
d3k crawl # Discover app URLs
d3k crawl --depth all # Exhaustive crawl
d3k agent-browser auto-connects to the active session's browser via CDP:
d3k agent-browser open http://localhost:3000/page
d3k agent-browser snapshot -i # Get element refs (@e1, @e2)
d3k agent-browser click @e2
d3k agent-browser fill @e3 "text"
d3k agent-browser screenshot /tmp/shot.png
To target a different browser, run d3k agent-browser connect <port> first.
Use this workflow when the user asks Codex to start d3k with a fresh browser/profile.
Close any stale agent-browser daemon before launching with --profile. Otherwise agent-browser will reuse the existing daemon and print --profile ignored.
d3k agent-browser close --all
Start the app through d3k in servers-only mode and keep that command running. In Codex, this is more reliable than asking d3k to launch the browser itself when a fresh profile is required.
d3k --no-agent --no-skills --servers-only --command "npm run dev -- -H 127.0.0.1 -p 3000" --port 3000 --startup-timeout 90 --no-tui
Adjust the package-manager command and port for the project. Prefer --command over --script when passing framework flags. For npm scripts, put flags after --; otherwise tools like Next.js can interpret the port as a project directory.
Verify the server before opening more browser windows:
curl -I http://127.0.0.1:3000
Open the fresh profile as a separate browser step:
d3k agent-browser --profile /tmp/d3k-fresh-profile --headed open http://127.0.0.1:3000
Sanity-check the opened page:
d3k agent-browser get title
d3k agent-browser snapshot -i
d3k errors
Practical rules:
127.0.0.1 for this workflow. If localhost hangs or flips between IPv4/IPv6 behavior, do not keep retrying browser launches.curl -I hangs, the server is wedged even if the port appears occupied; restart the d3k server process before opening a browser.servers-only mode there is no d3k-monitored CDP browser. Use regular d3k agent-browser commands, not d3k cdp-port.agent-browser opens outside the sandbox when sandbox networking blocks access to 127.0.0.1.Use agent-browser for browser work.
Practical rule:
agent-browser.d3k agent-browser snapshot -i
d3k agent-browser click @e2
To make d3k prefer one locally when it launches helper browser commands, use:
d3k --browser-tool agent-browser
d3k errors --context - See errors and what triggered themd3k agent-browser open <url> then d3k agent-browser click @e1 to replayd3k errors - Verify fix workedWhen creating a PR for visual changes, always capture before/after screenshots to show the impact:
Before making changes, screenshot the production site:
d3k agent-browser open https://production-url.com/affected-page
d3k agent-browser screenshot /tmp/before.png
After making changes, screenshot localhost:
d3k agent-browser open http://localhost:3000/affected-page
d3k agent-browser screenshot /tmp/after.png
Or use the tooling API to capture multiple routes at once:
capture_before_after_screenshots(
productionUrl: "https://myapp.vercel.app",
routes: ["/", "/about", "/contact"]
)
Include in PR description using markdown:
### Visual Comparison
| Route | Before | After |
|-------|--------|-------|
| `/` |  |  |
Upload screenshots by dragging them into the GitHub PR description.
Author a completed-spec fixture — a seeded intent graph plus a believable four-phase interview transcript — for a brunch feature, so it can feed any downstream tooling that consumes a completed spec or dogfood the UI. Use when you need a real completed specification in .brunch/brunch.db for testing or demo: pick a feature from memory/SPEC.md (or a free description), choose a stress profile (coverage|clean), and produce a registered scenario seed + an interview-questions doc that match exactly.
Review someone else's PR through an architectural/strategic lens — does it shift design, raise complexity, or quietly change precedent? Drives a brief reflective discussion with the user, then posts a concise review comment via `gh`. Defers line-level and bug-hunting concerns to AI review bots and the PR author.
Crystallize shared understanding into a reviewable spec, or update an existing one. Use when the problem needs a written specification, when assumptions or decisions have changed, or when the user says 'write a spec'.