# Sponsor Sizer

> An influencer and creator partnership workbench for founders and marketers: paste what you know about one creator (their platform, follower count, likes, comments and views on recent posts, a media kit, a rate card, captions of past sponsored posts, sample comments) and what you sell (the product, who buys it, the goal, budget, proof, competitors, a promo code or landing page, redlines), and find whether their audience fits, what the deal should be and what the agreement must lock, then write the creative brief with the disclosure rules built in. The free scan runs in the browser and uploads nothing. The paid vetting says partner, test first or pass. The paid brief lane writes what the creator receives.

App: https://sponsor-sizer.skillsafe.ai/
API tutorial: https://sponsor-sizer.skillsafe.ai/api.html

## Input

- The creator: plain text, Markdown, CSV, JSON or a saved HTML media kit, pasted or dropped. Handle, platform, follower / subscriber / listener count, per-post likes, comments and views (an average or a few posts), a stated engagement rate, a rate card, past sponsorships and their results, captions of past sponsored posts, sample comments, audience demographics.
- What you sell (optional but needed for fit, the brief and tracking): the product, the buyer, the goal, the budget, proof (reviews, results), competitors, a promo code, a UTM link or landing page, and what must never be said.
- Optional: the goal - awareness, conversions, content to reuse as ads, or trust.

## The free scan

- The numbers: platform (the one the follower count belongs to), handle, followers, tier (nano 1k-10k, micro 10k-50k, mid 50k-500k, macro 500k+, or a B2B voice for LinkedIn), likes / comments / views per post, the engagement rate as (likes + comments) / followers with the arithmetic shown, a stated engagement rate set against it, the quoted rate against the skill's starting single-post range for the tier (nano free product to $100, micro $100 to $1,500, mid $1,500 to $10,000, macro $10,000 to $30,000+; YouTube, podcasts and newsletters priced on cost per thousand instead), cost per thousand views or downloads, cost per engagement, and the budget.
- The five vetting checks from the skill, graded met / unclear / not met with the lines that carry them: engagement rate (a rough 1% floor at scale, 2% for nano), the fake-follower check (a follower spike, three or more generic comments, a comment pod, engagement out of line with reach, suspiciously round numbers), the sponsored track record (past sponsorships, and results from them), brand safety (a competitor you listed named in their notes, brand-safety words, or your note that you scrolled their last three months), and audience fit (always left to the paid vetting).
- Disclosure in each past promotional caption: clear (#ad, #sponsored, "Paid partnership" or "Ad:" in the first 125 characters), buried (below the fold or in a pile of hashtags), insufficient (#sp, #spon, #collab, #ambassador, #partner or "thanks to" alone) or missing.
- The agreement terms, each mentioned or not yet: deliverables, usage rights (whitelisting for a 3-6 month window), exclusivity, approvals, disclosure, payment, a kill / rework clause; and the tracking ready (promo code, UTM, vanity URL, a post-purchase survey).
- Flags CD-01 to CD-19: low engagement, a follower spike, generic comments or a pod, engagement out of line with reach, a quote above the tier's range, a sponsored post with no disclosure, insufficient or buried disclosure, a competitor conflict, brand-safety words, no audience information, no follower count, no per-post numbers, no tracking planned, no proof to brief from, usage rights not mentioned, round numbers, a budget below the tier's range, and a request for a word-for-word script.
- Exports: the scan as Markdown, a deal checklist, the numbers as CSV.

## The paid lanes

Both run on gpt-terra with a 10% publisher markup; /estimate is free and shows the credits reserved before a run.

- vet: the creator restated; partner, test first or pass; the tier; audience fit (strong, partial, weak, unknown) with quotes from both sides; the engagement read (healthy, low, suspicious, unknown); red flags with severity and quotes; the compensation model (flat, hybrid, performance, gifting, ambassador, ugc); the rate (quoted, the starting range, an anchor, judged on cost per qualified outcome); the agreement terms (always deliverables, usage rights and disclosure; exclusivity when you list competitors); the tracking to set up before launch; a 1:1 outreach opener naming a piece of their content; and what is missing.
- brief: the why in one sentence; two or three talking points, each backed by a quote from what you sell; one CTA with your code and link (or placeholders); guardrails (your redlines, no promised results, no implying they are a customer); the written disclosure (#ad or #sponsored at the start), the spoken line for video and audio, the platform's paid-partnership label; deliverables; usage rights; tracking; explicit creative freedom; the brief text to send (350 words or fewer, never a script); what to check before sending.
- Handoff: a vetting hands its model, terms and tracking to the brief lane in one click.

## What the page checks in every reply

Vetting: the tier against the page's follower count; partner only with a strong fit, healthy engagement and no high red flag; a pass when engagement is under 1% at 10,000 followers or more; the engagement read against the page's arithmetic; terms that cover deliverables, usage rights and disclosure (and exclusivity with listed competitors); tracking present; promo codes copied from the notes or left as placeholders; every quote in the notes word for word; no number the notes do not contain. Brief: two or three talking points each backed by a quote from what you sell; #ad or #sponsored in the written disclosure, never #sp, #collab or #ambassador alone; a spoken line for video and audio; the CTA's code and link yours or placeholders; at least two guardrails; the brief under 350 words with the disclosure in it; the vetting's model kept; no invented figures.

## Limits

The scan reads what the notes say. It cannot see a follower graph, a comment section or the last three months of posts. Rate ranges are the skill's negotiation anchors, not prices. This is not legal advice: regulated categories (health, finance, alcohol) and cross-border campaigns go to legal.

## Sources

- Derived from the agent skill @coreyhaines31/influencer-marketing by Corey Haines (github.com/coreyhaines31/marketingskills, MIT): https://skillsafe.ai/skill/@coreyhaines31/influencer-marketing - its influencer-to-ambassador spectrum, creator tiers and starting rate ranges, vetting checklist, compensation models and deal terms, disclosure rules, creative-brief structure, measurement rules and run-of-show checklist.
- Built on SkillSafe apps: https://skillsafe.ai/