Interview Signal Ledger — Count People, Not Quotes
Check interview quotes, count distinct participants, and separate prompted interest from counterevidence before making product decisions.
$12
Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+22 moreClaude CodeCursorCodex CLIMuseDots
OpenClaw
Manus
Grok
Grok Bot+18 agentsInterview Signal Ledger — Count People, Not Quotes
Example session with this skill installed
Check examples/ledger.json and explain the difference between sessions and participants. Draft a short decision memo that preserves counterevidence and prompted interest. Do not claim market validation or actual payment.
- Read your context and instructions
- Compiled the interview signal ledger
Synthetic four-session, three-participant example
Complete captured output from the bundled ledger checker follows. CHECKED concerns exact linkage and declared counts; it does not validate theme meaning or customer demand.
{
"status": "CHECKED",
"exit_code": 0,
"distinct_participants": 3,
"sessions": 4,
"source_hashes": {
"I1": "407a57e6784927999d37f189435d154a453ce9624b5fff8a7581fd0e4d1c1f1c",
"I2": "97d354a00bb40551594b5cffeb343f4ced18a0099e4f5962a3ac003ab91f71f1",
"I3": "3222c8c6f9d73ba45de2087ddd2a8730a5cefdf538c5534ab758772b09fd309a",
"I4": "04d638739943ec9c99d8a4168a8a81e98adaf7aa6c968dff5dc23e45b3392282"
},
"themes": [
{
"id": "T1",
"question": "Q1",
"eligible_participants": [
"P1",
"P2",
"P3"
],
"support_participants": [
"P1"
],
"counter_participants": [
"P2"
],
"prompted_support_participants": [
"P3"
],
"contradictory_participants": [],
"unprompted_support_count": 1,
"eligible_count": 3,
"links": [
{
"evidence_record": 1,
"source": "I1",
"linked": true,
"issue": null
},
{
"evidence_record": 2,
"source": "I2",
"linked": true,
"issue": null
},
{
"evidence_record": 3,
"source": "I3",
"linked": true,
"issue": null
},
{
"evidence_record": 4,
"source": "I4",
"linked": true,
"issue": null
}
]
}
],
"findings": [],
"caveat": "Exact quote linkage and declared participant counts only. Theme meaning, question/exposure labels, independent purchasing decisions and demand remain unverified.",
"input_sha256": "1b65b2464b3e156325aee97358c93991c759f86b14ff51d5b3a58b589d1c5637"
}
Worked interpretation
There are four sessions but only three distinct participants. P1 supports the theme in two sessions and still counts once. P2 supplies counterevidence. P3 supplies prompted hypothetical support, kept separate from the one unprompted supporting participant. All four declared quote links match the supplied source text.
Decision
Treat this as a question to investigate, not market validation. Review the actual passages, question wording and recruitment before interpreting the theme. A useful next research step is to examine current behavior and actual purchasing decisions with additional independent participants. No payment, conversion rate or statistical prevalence is established.
Connects securely to your tools. The creator never sees your data.
About this skill
Four interviews are not always four customers. Two sessions with the same person can make a weak signal look bigger, and “yes, maybe” after a leading question is not a paid order.
Interview Signal Ledger helps your agent synthesize the evidence without inflating it. Keep participant IDs separate from sessions, link themes to actual passages and distinguish observed behavior from stated intentions.
What you get
- A focused synthesis workflow and a decision memo tied to the supplied material.
- A standard-library Python checker for exact quote linkage, IDs and declared question coverage.
- Distinct participant counts for unprompted support, counterevidence and prompted support.
- Visible within-person contradictions instead of an averaged “positive” result.
- A synthetic four-session, three-participant ledger and a worked memo.
In the included example, one person supports the theme across two sessions, one person contradicts it, and another expresses prompted hypothetical interest. The correct unprompted support count remains one. Repeating their quotes does not create more customers; an invented quote fails the exact-source check.
No extra API or research platform is required. The checker reads the selected local JSON and outputs a report with hashes, without uploading it or modifying it.
Scope and limitations
Exact text matching does not prove the quote supports the theme, that question labels are truthful or that participants represent independent buying decisions. A carefully selected fragment can still mislead. The host/model and human reviewer must inspect interpretation, leading questions and recruitment bias.
CHECKED means declared mechanical links passed, not that your idea has demand. No recording, transcription, participant outreach, consent verification, statistical prevalence, willingness-to-pay proof or customer payment is inferred. You must supply anonymized identities and actual source text. Other agents and Agensi script-resource delivery have not been independently validated.
How to use
Skills on Agensi run inside your AI agent. Copy the prompt, paste it into your agent, and it connects to Agensi and installs this skill for you. If a payment is needed, your agent shows it to you first.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill passes our trust checks below.
Security scanned
Passed our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scanned
Verified clean today
- Passed all security checks, Safe to install
Requires