sales-skills/sales-frill

Frill Platform Help

Step 1 — Gather context

If references/learnings.md exists, read it first for accumulated platform knowledge.

1. What are you trying to do?

   • A) Read/sync ideas, votes, comments via the REST API (cursor pagination)
   • B) Verify a signed webhook (HMAC-SHA256) and react to idea/vote/comment events
   • C) Embed the widget / identify logged-in users / run a survey (Frill-Script)
   • D) Map Statuses + Topics to your public roadmap and changelog (Announcements)
   • E) Set up SSO or a native integration (Slack/Jira/Linear/Intercom/Zapier)
   • F) Decide Frill vs Canny/Nolt (price), or pick a plan

2. API or no-code? Code → REST API + signed webhooks. Front-end → Frill-Script widget. No endpoint → native integrations / Zapier.

Skip-ahead rule: if the user's prompt already provides enough context, skip to Step 2.

Step 2 — Route or answer directly

| If the question is about... | Route to... | |---|---| | Voice-of-customer / survey program strategy (NPS/CSAT) across tools | /sales-customer-feedback {question} | | Aggregating feedback across reviews/social/support sources | /sales-noisely {question} | | Connecting Frill to a CRM/PM tool generically (iPaaS) | /sales-integration {question} | | Public review generation strategy | /sales-customer-reviews {question} |

When routing, give the exact command, e.g. "This is a strategy question — run: /sales-customer-feedback build a closed-loop feedback program".

Step 3 — Frill platform reference

Read references/platform-guide.md for the full reference — the Ideas/Roadmap/Announcements/Surveys module map (what's API vs widget vs UI), the prioritization matrix, plan tiers, the data model, and quick-start recipes (pull ideas via the API; verify a webhook; embed + identify a user).

Read references/frill-api-reference.md for the integration surface — base https://api.frill.co/v1, Bearer API-key auth (from app.frill.co/settings/company), cursor pagination (after ← pagination.endCursor), the resources (ideas/comments/votes/statuses/topics/followers/announcements/notes), HMAC-SHA256 signed webhooks (events + triggered_at/event/data/data_type), Frill-Script (widget/surveys/identify), and SSO.

Answer using only the relevant section. Don't dump the full reference.

Step 4 — Actionable guidance

Focus on the user's specific situation:

• Bearer auth + cursor pagination. Authorization: Bearer <key> against api.frill.co/v1; page lists by passing pagination.endCursor as the next request's after param until hasNextPage is false. Keep the key server-side.
• Webhooks are HMAC-SHA256 signed — verify them. Compute HMAC-SHA256(rawBody, webhookSecret) (secret from Company Settings → Webhooks) and constant-time compare to the signature header. Payload is {triggered_at, event, data_type, data}; dedupe on the entity id + triggered_at.
• Statuses + Topics drive the roadmap. An Idea's Status is its roadmap column (suggestion → in progress → done) and Topics are tags; read/set them via the API to keep an external roadmap in sync.
• Identify users in the widget. Use Frill-Script to pass your logged-in user's identity so their votes/ideas attach to a real account (and SSO so they're auto-authenticated) — otherwise feedback is anonymous.
• It's a feedback-board tool, not VoC surveys or aggregation. Frill collects solicited board feedback + roadmap + changelog. For NPS/CSAT programs use /sales-customer-feedback; for mining unsolicited feedback across reviews/social use /sales-noisely.
• Price is the pitch. Frill positions as a cheap Canny/Nolt alternative (~$25/mo vs Canny's $400) — size the plan to seats/boards; startup/nonprofit discounts exist.

If you discover a gotcha, workaround, or tip not covered in references/learnings.md, append it there.

Gotchas

Best-effort from research (2026-06) — API verbatim from developers.frill.co; pricing from reviews. Confirm in-account.

1. Cursor pagination, not page numbers. Pass endCursor as after; loop until hasNextPage is false. Don't assume offset/page params.
2. Webhooks are HMAC-SHA256 signed — don't skip verification. Secret from Company Settings → Webhooks; compute over the raw JSON body and constant-time compare to the header.
3. Bearer header is the recommended auth (query-string ?api_key= leaks via logs; Basic auth uses the key as the username). Keep the key server-side.
4. Anonymous unless you identify users. Without Frill-Script user identification (or SSO), board votes/ideas aren't tied to a real account — wire identification if you need per-user data.
5. Feedback-board ≠ VoC survey ≠ aggregation. Frill is solicited board feedback + roadmap + changelog; it doesn't run NPS/CSAT programs (→ /sales-customer-feedback) or mine reviews/social (→ /sales-noisely).
6. Plan gates. Higher tiers unlock more boards/admins/white-label/integrations; the headline ~$25/mo is the entry point (Enterprise ~$349). Verify what your tier includes.
7. Some create/update fields aren't on the overview docs. The GitBook docs support .md + a ?ask= dynamic query — use them to confirm exact request bodies per resource.

Related skills

• /sales-customer-feedback — Voice-of-customer / survey program strategy across tools (Frill is a feedback-board option) — program design, tool selection
• /sales-featureos — FeatureOS (formerly Hellonext) platform help — the closest like-for-like feedback-board competitor (boards + roadmap + changelog + knowledge base); differs on API auth (API-KEY header + page/per_page vs Frill's Bearer + cursor) and unsigned dashboard webhooks. Install: npx skills add sales-skills/sales --skill sales-featureos
• /sales-sleekplan — AI feedback board (auto-categorization/duplicate merging) with an MCP server; the AI-forward affordable alternative to Frill. Install: npx skills add sales-skills/sales --skill sales-sleekplan
• /sales-userjot — UserJot platform help — AI-first feedback board + MCP server, no per-seat pricing (Canny/Productboard alternative); contrast with Frill on webhook signing (Frill is HMAC-SHA256-signed + cursor pagination vs UserJot's unpublished scheme) and the /requests post resource. Install: npx skills add sales-skills/sales --skill sales-userjot
• /sales-noisely — AI feedback aggregation across reviews/social/support (complement: Frill = solicited boards, Noisely = unsolicited mining)
• /sales-customer-reviews — Public review generation strategy
• /sales-integration — Connecting Frill to a CRM/PM tool via API/webhooks/Zapier
• /sales-do — Not sure which skill to use? The router matches any sales objective to the right skill. Install: npx skills add sales-skills/sales --skill sales-do -a claude-code

Examples

Example 1: Pull ideas + votes into my system via the API (developer/automation)

User says: "How do I sync all the feedback ideas and their vote counts out of Frill?" Skill does: Shows GET https://api.frill.co/v1/ideas with Authorization: Bearer <key>, paging via after= ← pagination.endCursor until hasNextPage is false, and reading each idea's status/topics/vote count (with /v1/votes / /v1/comments for detail). Notes keeping the key server-side, and that the webhook (idea/vote events) is better than polling for ongoing sync. Result: A complete, paginated export of ideas + engagement.

Example 2: Verify a Frill webhook so I can trust it

User says: "Frill is POSTing events to my endpoint — how do I verify them?" Skill does: Explains the HMAC-SHA256 signature: get the Webhook Secret (Company Settings → Webhooks), compute hash_hmac('sha256', rawBody, secret), and constant-time compare to the signature header (reject on mismatch). Notes the payload shape {triggered_at, event, data_type, data}, deduping on entity id + triggered_at, and which events exist (idea/vote/comment/announcement). Result: Authenticated, tamper-evident webhook intake.

Example 3: Frill vs Canny — which feedback tool?

User says: "We're a small SaaS — Frill or Canny for a feedback board + roadmap?" Skill does: Frames the tradeoff — Frill is the affordable pick (~$25/mo, boards + roadmap + changelog + surveys, REST API + signed webhooks), while Canny is pricier (~$400/mo lowest paid) with deeper enterprise features; Nolt is even simpler/per-board. Recommends Frill for cost-sensitive indie/SMB SaaS and routes broader tool selection: "run: /sales-customer-feedback choose a product-feedback board tool." Result: A price-grounded board-tool choice.

Troubleshooting

My API only returns the first page of ideas

Symptom: You get ~25 ideas and can't reach the rest. Cause: Frill uses cursor pagination — there are no page/offset params; you must follow the cursor. Solution: Read pagination.endCursor and pagination.hasNextPage from the response, then repeat the request with ?after=<endCursor> until hasNextPage is false. Accumulate results across pages.

My webhook signature check fails

Symptom: The HMAC you compute doesn't match Frill's signature header. Cause: Hashing a re-serialized body (not the raw bytes), the wrong Webhook Secret, or a wrong algorithm. Solution: Compute HMAC-SHA256 over the raw request body with the Webhook Secret from Company Settings → Webhooks, then constant-time compare to the signature header. Match the docs' hash_hmac('sha256', payloadJson, secret) exactly; once verified, dedupe on the entity id + triggered_at.

Votes/ideas in the widget are all anonymous

Symptom: Board activity isn't attributed to your actual users. Cause: The Frill-Script widget isn't identifying users (no SSO / no identity passed). Solution: Use Frill-Script user identification to pass your logged-in user's id/email/name (and/or set up SSO) so votes/ideas attach to real accounts. See developers.frill.co/frill-script/identifying-users. For VoC program strategy beyond boards, use /sales-customer-feedback.