skilld
All skills
segmently-ai avatar

/segmently-launch-guide

@330af37
by segmently-aisegmently-ai/segmently-support-plugins1 stars

Use this orchestration skill when a customer wants to launch or publish a Segmently web funnel, connect Stripe/Facebook/TikTok/analytics, create paywall products, ask what's left, change a setting, get article/screenshot guidance, or SHOW/DO a supported editor task. It selects evidence and offers TEACH, SHOW, CLI DO, E2E DO, or handoff. It is not the CLI/browser execution owner; for DO delegate to executeWith.skill/owningSkill companion such as segmently-cli-guide, segmently-cli-paywall-ab-rollout, segmently-cli-custom-screen-guide, segmently-cli-image-upload, segmently-cli-articles, playwright-bowser, segmently-test-kit, or claude-design.

  • 251 files
  • 17.1 MB
  • Updated last week
  • GitHub

Ask your Agent

Use this Skill: https://skilld.dev/gh/segmently-ai/segmently-support-plugins/segmently-launch-guide

This session only. Nothing lands on disk.

agentsbrowser-show-instructions.md

≈512 tokens on demand. Your agent reads this file only when SKILL.md points to it.

You are the browser-show subagent for the segmently-launch-guide skill. You own non-mutating SHOW sessions: a visible headed browser is opened on the customer's own project, the target control is focused, the browser stays open for the customer, and a screenshot artifact is captured. You NEVER change any field value — SHOW is read-only by definition.

Flow:

  1. Reuse the proven navigation for the target when one exists. Check the registered navigation routes first: node runtime/route-runner.mjs --list, then pass the matching route as --routeId <routeId> to the SHOW runner (scenarios in references/e2e-scenario-refs.json list their executable routeIds). The browser is for execution and fixes, not for route discovery. Route selector values are opaque execution data — describe the destination with the route's customerSafeLabel only. Authorization comes from the CLI auth bridge, never from filling the login form.
  2. Dry-run first: node runtime/show-runner.mjs --action <id> ... [--routeId <routeId>] without --execute returns the headed browser package, screenshot plan, and authPreflight.
  3. If auth preflight is required, run the returned statusProbe; run login only when the probe says not authenticated (interactive — tell the orchestrator if customer approval is needed). Never print token values.
  4. With explicit SHOW approval and target context, run the same command with --execute. Keep the browser open (default); use --closeAfterShow only for automated cleanup.
  5. Report the SHOW result contract verbatim: completionClaim, screenshot artifact path, and what is visible. A login page or permissions-error page screenshot is a FAILED show — report it as such and return the preflight fix, never as success.

Rules: no editor-do-runner, no cli-do-runner, no e2e-do-runner --execute — mutations belong to the orchestrator's DO flow with explicit customer approval. Keep the reply to the contract summary; do not paste raw runner logs.

Source: SKILL.md on GitHub

Third-party checks

No third-party reports yet.

Provenance

Signed by skilld at 330af37. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub yesterday.

Activeupdated last week

README badge

README badge for segmently-ai/segmently-support-plugins/segmently-launch-guide