influencer-discovery

Builds evidence-backed multi-platform influencer candidate pools with authenticity screening and non-ranked Fit-readiness queues.

Updated Jul 2, 2026
One-click install
npx skills add https://github.com/qaz26688442/bo-car --skill influencer-discovery-qaz26688442
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: influencer-discovery
Source: https://github.com/qaz26688442/bo-car/tree/main/.agents/skills/influencer-discovery
Command: npx skills add https://github.com/qaz26688442/bo-car --skill influencer-discovery-qaz26688442

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Finding credible influencers for a brand or niche requires searching multiple platforms, verifying authenticity, and organizing candidates with traceable evidence. This Skill turns a niche, platform, and follower-range request into a screened candidate pool with field-level evidence trails, red-flag screening, and a non-ranked readiness queue for downstream scoring. ## Core Features & Use Cases - Multi-Platform Candidate Discovery: Searches Instagram, TikTok, YouTube, X, LinkedIn, and Reddit via hashtags, similar accounts, competitor mentions, and optional connectors, with keyless oEmbed lookups and a free YouTube Data API connector for measured metrics. - Evidence-Profiled Screening: Filters candidates on follower range, engagement, recency, relevance, and brand safety, recording provider, source, observation date, window, and evidence label for every factual field. - Privacy-Safe Identity Handling: Uses opaque creator references, keeps raw handles and contact data transient, and triages candidates into READY_FOR_FIT, NEEDS_REFRESH, or INELIGIBLE queues without ranking. - Use Case: Ask for 15 micro-influencers (10K-100K followers) in sustainable fashion and receive a screened, evidence-backed candidate queue ready to hand off to fit-scorer for weighted ranking. ## Quick Start Find 20 influencers in the sustainable fashion niche with 50K-200K followers on TikTok and Instagram, engagement above 4%, for my eco clothing brand.

Frequently Asked Questions about influencer-discovery

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

FAQPage Schema
How do I find influencers in a specific niche for my brand?▼

Provide the brand or product, niche, target platforms, follower range, and engagement floor, and the Skill builds a screened candidate pool. It searches hashtags, similar accounts, competitor mentions, and platform-native discovery, then returns evidence-backed profiles in a non-ranked readiness queue.

What platforms does influencer discovery support?▼

It covers Instagram, TikTok, YouTube, X/Twitter, LinkedIn, and Reddit, with per-platform vetting playbooks for each. Keyless oEmbed endpoints resolve candidate metadata on YouTube, TikTok, and X, and a free YouTube Data API key upgrades YouTube metrics to measured values.

Does this Skill rank or score the influencers it finds?▼

No. It deliberately avoids scoring, tiering, or action-ranking candidates. It records evidence completeness and declared-filter results, then hands the queue to the fit-scorer Skill, which owns weighted comparison and ranking.

Can I use influencer discovery without paid tools or API keys?▼

Yes. Every step degrades to user-supplied inputs such as exports or public handles, and keyless oEmbed endpoints provide basic candidate metadata. Optional connectors like an influencer database or CRM sharpen results but are never required.

What happens if I only have a couple of handles and no criteria?▼

The Skill stops with NEEDS_INPUT and names the missing criteria rather than fabricating candidates. Two raw locators without complete criteria and evidence cannot become a vetted shortlist; only an explicitly authorized partial checkpoint may be saved.

How does the Skill handle creator privacy and contact data?▼

It uses opaque creator references and keeps raw handles, emails, and profile URLs transient only, never copying them into saved artifacts or handoffs. Contact paths are recorded as pseudonymous references, and outreach requires separate authorization after an approved campaign plan.