Update to august

{
  "name": "august-flow-audit",
  "description": "Walk August twins through a proposed onboarding, checkout, signup, or funnel step-by-step and surface friction points before we ship. Screenshots, Figma frames, or step descriptions are attached as focus group stimuli one step at a time; twins call out where they'd get confused, hesitate, or drop off. Pre-launch, twin-simulated version of the friction points Commerce Graph measures post-launch. Triggers on: \"audit this flow\", \"walkthrough with twins\", \"onboarding friction check\", \"test this funnel with twins\", \"friction check\", \"flow audit\", \"/august-flow-audit\".",
  "included_files": [],
  "skill_md_contents": "---\nname: august-flow-audit\ndescription: >\n  Walk August twins through a proposed onboarding, checkout, signup, or\n  funnel step-by-step and surface friction points before we ship.\n  Screenshots, Figma frames, or step descriptions are attached as focus\n  group stimuli one step at a time; twins call out where they'd get\n  confused, hesitate, or drop off. Pre-launch, twin-simulated version of\n  the friction points Commerce Graph measures post-launch. Triggers on:\n  \"audit this flow\", \"walkthrough with twins\", \"onboarding friction check\",\n  \"test this funnel with twins\", \"friction check\", \"flow audit\",\n  \"/august-flow-audit\".\n---\n\n# August Flow Audit\n\nWalk twins step-by-step through a proposed flow and produce a per-step\nfriction report plus a top-3 overall list. Tools under\nthe August MCP.\n\n**When to use:** the user has an ordered flow (onboarding, checkout,\nsignup, quiz, guided-shopping funnel, dashboard first-run) and wants\nfriction signal before shipping.\n\n**When NOT to use:**\n- Single screen or artifact → `/august-feedback`.\n- Comparing variants of the same screen → `/august-concept-compare`.\n- Text/microcopy only → `/august-feedback` for one version or\n  `/august-concept-compare` for multiple versions.\n- Already-shipped flow with live traffic → skip this and pull the real\n  friction points from Commerce Graph.\n\n**Core rule:** the flow order MATTERS. Twins react to step N with step\nN-1's context in their head. Never scramble the order and never merge\nsteps into one stimulus batch — the whole point is step-by-step signal.\n\n---\n\n## STEP 1 — Confirm the flow\n\nBefore any MCP call, be clear on:\n\n- **Ordered list of steps** — screenshots, Figma frame URLs, or written\n  descriptions. Cap at ~8 steps per audit; if longer, chunk into two\n  audits (e.g., \"onboarding\" and \"first-purchase\").\n- **What is this flow FOR?** (new-user onboarding, checkout, guided\n  quiz-to-recommendation, dashboard first-run, etc.)\n- **Who is the target user?** (first-time visitor, returning buyer,\n  internal dashboard user, etc.)\n- **What's the desired end state?** (completed purchase, activated\n  account, added product to cart, etc.)\n\nMissing any → ASK. A flow audit without a stated goal produces\ndirectionless friction complaints.\n\n---\n\n## STEP 2 — Pick the organization\n\nCall `list_organizations`. Map audience → org:\n\n- Consumer flow (checkout, funnel, onboarding on a brand site) → brand\n  org.\n- Internal flow (Fermat dashboard first-run, admin setup wizard, August\n  research-builder onboarding) → Fermat org.\n\nIf unclear, ASK.\n\n---\n\n## STEP 3 — Start the focus group\n\nUse a focus group, not single-twin — friction signal wants diversity. Call\n`list_focus_groups` for the org and select an existing ready group whose\nconfigured panel matches the target user. Then call `start_focus_group` with\nits `focus_group_id`. The MCP cannot create a group or choose its panel; if\nno suitable ready group exists, ask the user to configure one in August.\n\nKeep this focus group open across every step. Do NOT end it between\nrounds — the twins' running memory of prior steps IS the signal.\n\n---\n\n## STEP 4 — Run the walkthrough, step by step\n\nFor each step in order:\n\n1. For a Figma URL, attach it with `add_focus_group_stimuli`. For a screenshot\n   description or text step, include the full labelled artifact directly in\n   the round question; the stimuli tool accepts only Figma URLs.\n2. `run_focus_group_round` with a friction-focused prompt:\n\n   > This is step [N] of [total]: [step name]. Imagine you just came from\n   > the previous step. React honestly:\n   > - Where would you get confused here?\n   > - Where would you hesitate, second-guess, or want to go back?\n   > - Where might you drop off entirely?\n   > - What would you expect to happen next?\n\n3. Optionally, on complex steps, run a second `run_focus_group_round`\n   asking each twin to score friction 0–3 on that step (0 = smooth, 3 =\n   would abandon) — this gives a rough severity anchor.\n\nAfter the last step:\n\n- `end_focus_group`\n- `get_focus_group_results` for the full transcript across all rounds.\n\n---\n\n## STEP 5 — Synthesize into a friction report\n\nDo NOT dump transcripts. Return this template:\n\n```\n## Flow Audit: [flow name]\nAudience: [org] · Twins: [N] · Goal state: [end state]\n\n### Per-step friction report\n\n**Step 1 — [name]**\n- Top friction points:\n  - [point 1 — one line]\n  - [point 2 — one line]\n- Standout quote: \"[verbatim from a twin]\" — [twin name]\n- Severity: **[low / med / high]**\n- Suggested fix: [one line — concrete change, not vibes]\n\n**Step 2 — [name]**\n- ...\n\n(repeat per step)\n\n### Top 3 friction points overall\n1. **[step + issue]** — [why it matters + severity]\n2. **[step + issue]** — [...]\n3. **[step + issue]** — [...]\n\n### Cross-step patterns\n- [e.g., \"twins repeatedly forgot what step 2 asked them, suggesting\n  progress indicator is under-weight\" — patterns visible only across\n  steps, not within a single step]\n\n### Post-launch note\nWorth running a real Commerce Graph friction-points check on this flow\nafter ship — the twin signal above is directional, not measured. Compare\nthe top-3 twin frictions against actual drop-off funnels once live traffic\naccrues.\n\n### Source\nFocus group [name/id], org [name], twins [names]\n```\n\nSeverity guidance:\n- **High** — twin explicitly said they'd bounce, or multiple twins hit\n  the same wall.\n- **Med** — twin hesitated / had to re-read / needed to guess.\n- **Low** — nit or preference, not a blocker.\n\nIf any step's signal was thin (twins just said \"looks fine\"), SAY SO —\ndon't pad. Suggest re-running with a different twin composition or a more\nspecific prompt.\n\n---\n\n## Quick reference\n\n| Situation | Action |\n|---|---|\n| Ordered flow with 3–8 steps + clear goal | Proceed |\n| >8 steps | Chunk into 2 audits, run separately |\n| Single screen | Redirect to `/august-feedback` |\n| Already-shipped flow with traffic | Pull real Commerce Graph friction points instead |\n| Twin signal thin on a step | Note it; offer to re-run with different panel |\n"
}