testimonial-miner

Mines captured messages, reviews, and meeting transcripts into a permission-tiered testimonial quote bank.

84|37|Updated May 23, 2026
One-click install
npx skills add https://github.com/legioncodeinc/vibe-coding-tools --skill testimonial-miner-legioncodeinc
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: testimonial-miner
Source: https://github.com/legioncodeinc/vibe-coding-tools/tree/main/src/quarantine/skills/keep/testimonial-miner
Command: npx skills add https://github.com/legioncodeinc/vibe-coding-tools --skill testimonial-miner-legioncodeinc

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Praise from clients arrives scattered across DMs, public comments, reviews, and meeting calls, and none of it is findable when a launch page needs social proof. Worse, quoting a private message or an unsubstantiated results claim without permission creates legal exposure under right-of-publicity law and FTC endorsement rules. ## Core Features & Use Cases - Praise discovery sweep: Runs narrow parallel queries across Littlebird MCP messages, screen snapshots, summaries, and meeting tools to find gratitude, results, and referral language in ten distinct registers. - Permission tiering: Assigns every quote a tier from its source channel (public and reusable, private and needs permission, or confidential and do not use), with unknown channels defaulting to confidential. - FTC results-claim flags: Separates results quotes into a sub-bank flagged SUBSTANTIATED, NEEDS EXPECTED-RESULTS DISCLOSURE, or NOT PUBLISHABLE AS A RESULTS CLAIM. - Use case: Before a product launch, run a deep sweep over 180 days to produce a persistent testimonial-bank.md with verified attribution, drafted permission requests held for approval, and a gap report naming client relationships with no captured praise. ## Quick Start Ask the assistant to find your testimonials and build a testimonial bank from your captured messages, reviews, and meetings.

Frequently Asked Questions about testimonial-miner

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find testimonials from my client messages and meetings?▼

Run a deep sweep that queries captured messages, screen snapshots, and meeting summaries across ten praise registers over a 180-day window. Each candidate quote is verified for speaker, target, exact wording, date, and role before entering the bank.

Do I need permission to quote a client in my marketing?▼

Public quotes from reviews or public posts are usable with attribution and a courtesy heads-up. Private quotes from DMs or emails require explicit written permission first, and confidential material from calls or NDAs should never be used or requested.

What tools does this skill require to run?▼

It requires the Littlebird MCP connected on a Power or Pro plan, using search_user_context, meeting search and retrieval tools, and routine management tools. If no Littlebird tools are present in the session, the skill stops rather than attempting a partial run.

Can I publish a client quote that mentions a specific result?▼

Results quotes are flagged into a separate sub-bank because the FTC reads them as representing typical outcomes. You must either substantiate typicality, disclose generally expected results clearly, or avoid publishing it as a results claim.

Can the skill edit or improve a customer quote?▼

No. Only rule-numbered trims for length are allowed, with the original stored alongside. Rewriting a customer's words is treated as fabrication and a potential FTC violation, since AI-generated review text is a prohibited category.

What are the limitations of automated testimonial mining?▼

It never assesses whether a result is typical, since a praise bank only contains happy clients. It also treats screen capture as showing what was viewed rather than written, so ambiguous attribution is dropped or escalated rather than guessed.