Voicepanel (voicepanel.com) platform help — a YC-backed AI-moderated customer-research platform that interviews REAL people across voice, video, chat, and phone, with screen-share usability tasks and stimuli testing (prototypes, sites, AI outputs), on a 30M+ global panel or your own users, adaptive probing, and Slack-native AI themes. Its automation surface is MCP-first — an MCP server (Claude/Cursor/ChatGPT over HTTP+OAuth) exposing studies, distributions, responses, transcripts, and analysis — with no public REST API or webhooks, and MCP access onboarded in waves. Use when setting up a Voicepanel study or usability task, recruiting panel vs your own users, wiring the MCP server into Claude Code, getting data out without a REST API, reading its sales-gated pricing, or choosing Voicepanel vs User Intuition / Strella / Outset. Do NOT use for the validate-before-building method or comparing tools broadly (use /sales-idea-validation), or analyzing existing NPS/CSAT/VoC feedback (use /sales-customer-feedback).
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
The command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
File Explorer
4 files
Showing SKILL.md
SKILL.md
Source instructions · Read-only preview
name
sales-voicepanel
description
Voicepanel (voicepanel.com) platform help — a YC-backed AI-moderated customer-research platform that interviews REAL people across voice, video, chat, and phone, with screen-share usability tasks and stimuli testing (prototypes, sites, AI outputs), on a 30M+ global panel or your own users, adaptive probing, and Slack-native AI themes. Its automation surface is MCP-first — an MCP server (Claude/Cursor/ChatGPT over HTTP+OAuth) exposing studies, distributions, responses, transcripts, and analysis — with no public REST API or webhooks, and MCP access onboarded in waves. Use when setting up a Voicepanel study or usability task, recruiting panel vs your own users, wiring the MCP server into Claude Code, getting data out without a REST API, reading its sales-gated pricing, or choosing Voicepanel vs User Intuition / Strella / Outset. Do NOT use for the validate-before-building method or comparing tools broadly (use /sales-idea-validation), or analyzing existing NPS/CSAT/VoC feedback (use /sales-customer-feedback).
argument-hint
[describe what you need help with in Voicepanel]
license
MIT
version
1.0.0
tags
["sales","pre-launch","platform"]
Voicepanel Platform Help
Voicepanel (voicepanel.com) runs AI-moderated customer research on REAL people — a YC-backed
platform whose tagline is "AI builds. Humans decide." An AI moderator runs interviews across
voice, video, chat, and phone, with adaptive real-time probing, screen-share usability
tasks, and embedded stimuli (prototypes, websites, videos, apps, AI outputs). Recruit from a
30M+ consumer/professional panel (150+ countries) or bring your own users (share link,
third-party redirect, in-product intercept), with unlimited screening/quotas. Results come back as
quantified themes with evidence and Slack-native AI summaries with audio/video clips, across
35+ languages with automatic translation.
Its niche in the AI-moderated-interview cluster is being broad-format and MCP-first: it adds
phone interviews and screen-share usability tasks most peers don't, and its automation surface is
an MCP server rather than a public REST API. That MCP-first posture is the thing to get right —
see below.
Two things to say almost every time:
It's real people, but a stated "I'd pay" is still not a purchase. Real interviews beat synthetic
signal (genuine reasoning, objections, language) — but the willingness-to-pay go/no-go belongs to a
real behavior test (pre-sale, smoke test), not a moderator transcript. Route that to /sales-idea-validation.
Automation is MCP-first, not REST. Voicepanel exposes an MCP server (OAuth over HTTP) but
no documented public REST API or webhooks, and MCP access is onboarded in waves (request it;
it's Enterprise-associated). So "call the Voicepanel API / fire a webhook on completion" is the wrong
mental model — the agent-native path is the MCP server; the fallback is manual export.
Step 1 — Gather context
If references/learnings.md exists, read it first for accumulated platform knowledge.
Ask only what you can't infer:
What do you want from Voicepanel?
A) Set up a study — objective, questions, pick voice vs video vs chat vs phone, add stimuli or a screen-share usability task
B) Recruit — the 30M+ panel (with screening/quotas) vs bring-your-own users (share link, redirect, in-product intercept)
C) — themes, evidence clips, transcripts, segmentations, the Slack summary
Read results
D) Automate / get data out — the MCP server (for Claude/Cursor) vs manual export (no REST API/webhooks)
E) Choose — Voicepanel vs User Intuition / Strella / Outset / Listen Labs for your use case
Is this discovery (strangers on the panel) or feedback from your own users? That decides panel
vs bring-your-own and how honest the signal is.
Skip-ahead: if the user wants to compare research tools across the market or the
validate-before-building method, that's a /sales-idea-validation question — route in Step 2.
Step 2 — Route or answer directly
If the user's question is about…
Route to
Comparing research/idea/synthetic tools across the market, or the validate-before-building method
/sales-idea-validation {question}
A REST API / webhook-native interview pipeline (Voicepanel has none — it's MCP-first)
/sales-userintuition {question}
Running a real behavior demand test (smoke-test page, waitlist, pre-sale) after interviews
/sales-idea-validation or /sales-funnel{question}
Analyzing existing NPS/CSAT/VoC/review feedback (post-launch, not new interviews)
/sales-customer-feedback {question}
A synthetic-persona peer (AI personas instead of real people)
/sales-syntheticusers or /sales-ditto{question}
When routing, give the exact command: "This is a {domain} question — run: /sales-idea-validation {original question}"
Otherwise, answer Voicepanel-specific questions using Step 3.
Step 3 — Voicepanel reference
Read references/platform-guide.md for the full reference — the module/automation-surface table
(what's MCP-accessible vs UI-only vs manual export), best-effort pricing and plan gates, the
Study → Distribution → Response → Analysis data model, recruiting (panel vs bring-your-own), and the
usability-task workflow. Read references/voicepanel-api-reference.md for the MCP server surface
(auth, clients, the read/write tool list, the access gate).
Answer using only the relevant section — don't dump the full reference.
Step 4 — Actionable guidance
Say the caveat: a real interview is not a purchase. Whatever the user asks, make explicit that
Voicepanel interviews real people (a big step up from synthetic personas — genuine reasoning,
objections, language), but a stated "I'd pay" in an interview is not observed demand. Keep the
why/objections/language and take the go/no-go from a real behavior test (smoke-test click,
pre-sale) — route that to /sales-idea-validation.
Treat automation as MCP-first, and say the MCP access is gated. When any automation/export/
integration comes up, state that Voicepanel's programmatic surface is an MCP server (OAuth over
HTTP; works with Claude, Cursor, ChatGPT, Windsurf), not a public REST API or webhooks — and that
MCP access is onboarded in waves, so the user likely must request access (contact form /
sales@voicepanel.com) and it is associated with higher/Enterprise tiers. Don't design a REST/webhook
pipeline; if a webhook-native pipeline is a hard requirement, route to /sales-userintuition.
When wiring the MCP server, mirror in-app roles and check the tool split. The agent only sees
what the account role already allows: read tools (studies, plans, distributions, responses +
transcripts, study stats, analysis questions, insights, segmentations) are available to all roles;
write tools (create/generate/edit studies, create/edit distributions, publish panel recruits,
update responses/distribution status, run analysis, request review, create segmentations) require an
editor/admin role. Tell the user to connect with a role scoped to what the agent should do — a
viewer token can't launch a paid recruit. Setup: add the server URL and complete the OAuth login.
Pick the modality to the question, and use usability tasks for prototypes/AI outputs. Voice/phone
for fast emotional depth, video when you need to see the person, chat for reach/async, and a
screen-share usability task when you're testing a prototype, website, or an AI feature's output
("watch them use it"). This screen-share/usability-task capability is a Voicepanel differentiator —
recommend it when the user is evaluating a product artifact, not just asking questions.
Panel vs bring-your-own: match to the question. The 30M+ panel (150+ countries, screening/
quotas) reaches strangers for discovery/concept tests; bring-your-own (share link, third-party
redirect, in-product intercept) interviews your own users for churn/win-loss and is the honest
audience for product feedback. In-product intercept is the low-friction way to catch real users mid-task.
Present all pricing as best-effort, and warn that public $99/free figures are stale. Voicepanel's
live pricing is sales-quoted (tiers roughly Pay-as-you-go → Pro (14-day trial) → Enterprise); the
$99/mo and free-50-responses numbers still on third-party aggregators are outdated and no longer
on the pricing page. State any figure as best-effort and point the user to voicepanel.com/pricing
or a sales conversation to confirm — even the "free trial" routes through sales.
If you discover a gotcha or tip not in references/learnings.md, append it there with today's date.
Gotchas
Best-effort from research (2026-07) — pricing (sales-quoted, stale $99/free figures on aggregators),
panel size ("30M+"), language count (35+; older marketing said 29), and the MCP tool surface / access
gate all move; verify at voicepanel.com and voicepanel.com/mcp.
Real ≠ demand. Interviews reveal the why better than any synthetic tool, but a stated intent to
pay is not a purchase — the go/no-go still belongs to a real behavior test (pre-sale, smoke test).
No public REST API or webhooks — automation is the MCP server. Don't design a webhook-on-completion
pipeline; the agent-native path is MCP, the fallback is manual export. A REST/webhook pipeline is a
/sales-userintuition job.
MCP access is onboarded in waves and gated. It's not instant self-serve — you request access
(contact form / sales@voicepanel.com), and MCP/API is associated with Enterprise (the Agent tier
doesn't list it). Don't promise an immediate connection.
MCP respects your in-app role. The agent can't exceed the connected account's permissions — a
viewer is read-only and can't launch a paid panel recruit; use an editor/admin token for write actions.
Pricing is sales-gated and the public numbers are stale. The $99/mo tier and free-50-responses
plan on aggregators are outdated; the live page routes every plan to sales. Treat all figures as
best-effort and confirm on site.
A "free trial" still goes through sales. The 14-day Pro trial is not a self-serve signup — it's a
sales-mediated conversation. Set expectations accordingly for solo makers.
Panel recruiting spends money per respondent. Launching a panel recruit incurs usage-based
recruiting cost; screen tightly and pilot a small batch before scaling — and note a write-capable MCP
agent can trigger that spend.
Related skills
/sales-idea-validation — The tool-agnostic validate-before-building method + the full research/persona/validator tool landscape (use this to actually decide build-or-not; an interview "yes" is not demand). Install: npx skills add sales-skills/sales --skill sales-idea-validation -a claude-code
/sales-userintuition — The REST-API/webhook/MCP real-human interview peer — the pick when you need a webhook-native pipeline Voicepanel's MCP-only surface can't offer. Install: npx skills add sales-skills/sales --skill sales-userintuition -a claude-code
/sales-strella — The deterministic-scripted-flow, UI-only real-human interview peer (no automation surface at all — Voicepanel's contrast is the MCP server). Install: npx skills add sales-skills/sales --skill sales-strella -a claude-code
/sales-chikka — The self-serve, marketing-forward real-human voice-interview peer (free tier + public pricing, vs Voicepanel's sales-quoted plans) — UI-only (no API/MCP), but spins interviews into VoiceVerify testimonials. Install: npx skills add sales-skills/sales --skill sales-chikka -a claude-code
/sales-customer-feedback — Real-customer NPS/CSAT/VoC strategy across tools (use to analyze existing feedback; Voicepanel proactively runs new interviews and usability tasks). Install: npx skills add sales-skills/sales --skill sales-customer-feedback -a claude-code
/sales-funnel — Build the smoke-test / fake-door landing page that turns an interview hypothesis into a real demand test. Install: npx skills add sales-skills/sales --skill sales-funnel -a claude-code
/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: "The interviews loved my prototype — should I build it?"
User says: "I ran 25 Voicepanel interviews and most said they'd use and pay for it. Green light?"
Skill does: Credits the signal as real (actual people, genuine reasoning — far better than a
synthetic tool), then draws the line: a stated "I'd pay" in an interview is not a purchase. Tells
the user to mine the evidence clips and transcripts for objections, hesitation, and pricing
language (the real value — not the top-line theme frequency), then take the go/no-go from a real
behavior test (pre-sale or a "buy" click) via /sales-idea-validation. Flags pricing as best-effort.
Result: The founder keeps the qualitative depth and runs a pre-sale before committing weeks of build.
Example 2: "How do I pull Voicepanel results into Claude Code / my stack?" (developer/automation)
User says: "I want my agent to read completed interview transcripts and analysis programmatically."
Skill does: States that Voicepanel's programmatic surface is an MCP server (OAuth over HTTP;
works with Claude/Cursor/ChatGPT), not a public REST API or webhooks — so the agent reads
studies, distributions, responses + transcripts, analysis, insights, segmentations via MCP tools,
scoped to the connected account's role. Notes MCP access is onboarded in waves (request it;
Enterprise-associated) and that setup is "add the server URL + complete the OAuth login." If the user
needs a webhook-on-completion pipeline, routes to /sales-userintuition as the REST/webhook-native
alternative.
Result: The user requests MCP access and wires the agent instead of hunting for a REST endpoint that doesn't exist.
Example 3: "What does Voicepanel cost, and can I just sign up for the free plan?"
User says: "I saw a $99/mo plan and a free tier — I'll start there."
Skill does: Warns that the $99/mo and free-50-responses figures on aggregators are stale — the
live pricing page routes every plan to sales (tiers roughly Pay-as-you-go → Pro with a 14-day trial
→ Enterprise), and even the free trial goes through a sales conversation, not a self-serve signup.
Presents all numbers as best-effort, points to voicepanel.com/pricing, and — for a solo maker who
wants instant self-serve — notes User Intuition's self-serve free interviews as a contrast via
/sales-idea-validation.
Result: The user sets correct expectations (sales-gated) instead of expecting an instant free signup.
Troubleshooting
"Where's the Voicepanel REST API / webhook to sync interviews?"
Symptom: You're looking for an API key and a webhook to fire on interview completion.
Cause: Voicepanel has no documented public REST API or webhooks — its programmatic surface is an MCP server (OAuth over HTTP), and MCP access is onboarded in waves.
Solution: Request MCP access (contact form / sales@voicepanel.com), connect from Claude/Cursor by adding the server URL and completing the OAuth login, and read results with the MCP read tools. If a webhook-on-completion pipeline is a hard requirement, use an API-native tool instead — route to /sales-userintuition (REST API + HMAC-signed webhooks + MCP).
"My agent can read studies but can't launch a recruit"
Symptom: The MCP connection reads fine but write actions (create study, publish recruit) fail.
Cause: MCP mirrors your in-app role — read tools work for all roles, but write tools (create/edit studies, publish panel recruits, run analysis) require editor/admin; a viewer token is read-only.
Solution: Reconnect with an editor/admin-scoped account (or have an admin grant the role). A write-capable agent can trigger paid panel recruiting — scope the token to what the agent should be allowed to spend.
"The pricing I found doesn't match the site"
Symptom: Aggregators show a $99/mo plan and a free tier; the live site shows neither and asks you to contact sales.
Cause: Voicepanel moved to sales-quoted pricing; the $99/free figures on third-party sites are outdated.
Solution: Treat all published numbers as best-effort, confirm at voicepanel.com/pricing or via sales, and expect even the 14-day Pro trial to route through sales rather than a self-serve signup.
