Audit X (Twitter) influencers with AI agents
Score X creators in about 30 seconds. We pull 100 recent tweets, sample 100 followers for bot detection, and return a structured 0–100 score with red flags — before you send a single DM.
Four sub-scores. One verdict.
Each X audit returns four weighted sub-scores that combine into a 0–100 composite and an S/A/B/C tier.
Reach
40% weightLog-scaled follower count, median tweet impressions, and engagement-per-impression rate. Captures real visibility, not vanity follower counts.
Audience
30% weightFollower-to-following ratio, account-age vs posting velocity, and bot detection on a sampled follower set. Filters out paid or automated audiences.
Content
20% weightNiche concentration of the top 3 recurring topics across recent tweets. Specialists score higher than generalists for relevant brand fit.
Conversion
10% weightCTA cadence, brand mention frequency, and bio link presence — signals that the creator's audience actually clicks through to action.
Red flags we surface for X creators
- Sampled followers appear automated or recycled
- Engagement-to-impression ratio collapses under follower count
- Posting velocity suggests bot-assisted scheduling
- Feed reads as paid-promo heavy (high CTA rate)
- Locked or suspended account state
- Sudden follower spikes inconsistent with engagement growth
From X handle to verdict in 30 seconds
Three steps, one API call, no SDK to install.
Paste a X handle or link
Drop a creator's X (Twitter) profile URL or @handle in the dashboard — or hit /api/score from your agent.
We pull tweets and followers
100 recent tweets, a sample of followers, and an LLM pass over both — bot detection, niche focus, brand mentions, and red flags.
Return a structured 0–100 score
Four sub-scores (reach, audience, content, conversion), red flags, suggested CPM, and a shareable report URL. Same JSON shape as every other platform.
Built for X sponsorship workflows
Used by brand teams, agencies, and AI agents to vet creators before money changes hands.
Brand sponsorship teams
Vet a shortlist of X creators before reaching out. Save the report URL, share it with stakeholders, and skip the 50K-follower account with 200 likes per tweet.
Agencies running outreach at scale
Batch-audit hundreds of handles, filter to tier S/A, and only spend pitch time on creators with verified audience quality.
AI agents and workflows
Pipe audit JSON into Claude, Hermes, or your own pipeline. Auto-filter, auto-pitch, auto-decline — same structured response shape every time.
X audits, answered
We pull 100 recent tweets and a 100-follower sample via the X API, run an LLM analysis over both, and compute four weighted sub-scores (Reach 40%, Audience 30%, Content 20%, Conversion 10%). The composite rounds to 0–100 and maps to a tier S/A/B/C.
Yes. We sample 100 followers per audit and score authenticity using follower/following ratios, account age, posting velocity, and engagement signals. If a large share of the sample appears automated, that's surfaced as a red flag and lowers the Audience sub-score.
Each audit hits the X API live, so the data is as fresh as X allows. Reports cache by X user ID for 24 hours — if anyone else audits the same handle within that window, they get the cached report for free.
Private accounts can't be audited — we need to read recent tweets. Suspended and not-found accounts return a clear error without burning a credit. Locked accounts produce a low-confidence score with a red flag.
Because follower counts are gameable and don't predict ROI for brands. We weight engagement-per-impression, follower authenticity, and content focus heavily — so a 5K-follower niche X creator can outscore a 500K generalist with bought followers.
Vet creators across every channel
Same score model, same JSON shape — no matter which platform the creator is on.
Audit your first X creator today
Real X engagement signal in 30 seconds. Pick a plan that fits your volume.