Takes a post (pass a public post URL or an internal post id) plus a target niche/brand description, and produces an adapted content idea: rewritten title and description, segment-by-segment script, a short analysis of why the original went viral, and an execution checklist, the proven format transplanted into your niche. The URL does NOT need to be tracked: an unknown post is fetched and ingested automatically as part of the job, so any public TikTok/Instagram/YouTube link works with no crawl_profile round-trip. Asynchronous (~1–3 minutes, a few minutes longer for an untracked URL): returns a jobRef; poll get_remix_result until its "remix" field is populated (it returns status + result together). For video posts the remix auto-generates the transcript and visual analysis first, so quality no longer depends on you transcribing beforehand. Credits are refunded automatically if the remix fails.
Turn any public viral post into a content idea adapted to your own niche: a rewritten hook and title, a segment-by-segment script, a short analysis of why the original went viral, and an execution checklist that keep the proven structure while swapping in your topic, brand and audience. Hand it a post (by URL or id) and a description of your niche, and it does the format transplant for you; a URL the database has never seen is fetched and ingested automatically. It runs asynchronously, costs 20 credits (1 credit = $0.01), and refunds automatically if it fails.
Connect the Viral Outliers MCP server once and the remix_post tool becomes available to your agent alongside all other skills:
claude mcp add --transport http viral-outliers https://viraloutliers.com/api/mcp \
--header "Authorization: Bearer so_live_YOUR_KEY"In Claude.ai or ChatGPT, add a custom connector pointing at https://viraloutliers.com/api/mcp. Your agent then calls remix_post on its own whenever the task needs it.
| Name | Type | Description |
|---|---|---|
url | string | Public post URL. The easy path when a user hands you a link they saw. Does not need to be tracked: unknown posts are ingested automatically. |
postId | string | Internal post id (from search results). Provide this or url. |
targetNiche* | string | Your niche/brand/audience, 5–500 chars (e.g. "B2B SaaS founders selling analytics tools"). |
tone | string | Preferred tone of voice for the adaptation. |
The finished remix is a structured content idea: a rewritten title and description, a script broken into segments (hook, body beats, call to action), a short whyItWorks analysis of what made the original go viral, and an execution checklist for shooting or producing it. The winning post's structure is preserved; the topic, examples and voice are swapped to your niche, so you get a ready-to-produce brief rather than a vague "make something like this".
The call is asynchronous (roughly 1-3 minutes; a few minutes longer for a URL that was not tracked yet, since the post is fetched and analyzed first). It returns a jobRef immediately; poll get_remix_result until its remix field is populated (it returns status and result together), or watch get_job_status. The result also lands in the account's Content Ideas dashboard in the web app, so a human can pick it up there.
You supply the source post and a targetNiche (5-500 characters describing your brand and audience), plus an optional tone. The post reference can be a URL the database has never seen: the job fetches the post from the platform, stores it, and then remixes it, so there is no crawl_profile round-trip and no requirement that the creator is tracked. For a video post the remix auto-generates the transcript and visual scene analysis first, then adapts from that full understanding of the original, so you never have to transcribe beforehand; the trade-off is that a video remix takes a little longer than a text-post remix while that enrichment runs.
Because enrichment is cached on the post, remixing the same post again (or one you already transcribed) reuses the existing transcript and analysis instead of regenerating them. Credits are charged when the job is queued and refunded automatically if it fails, so a broken generation never costs you.
The flagship is a full content-research loop: search_outliers in a niche, pick the winners, and remix_post each into a client's niche to deliver ready scripts on a schedule. Format-transplant workflows find a small-account outlier (maxFollowers set low), then remix it into a new niche before the format saturates. Agencies run one proven post through several targetNiche descriptions to produce a brief per client from a single source.
Because the output is structured, downstream tools can act on it directly: feed the script segments into a teleprompter, a shot list, or a rendering pipeline without re-parsing prose.
You could paste a caption into a general-purpose model and ask it to "adapt this", but it would be working from the caption alone, blind to what actually happens in the video, and with no consistent structure to its output. The quality of an adaptation depends on understanding the original's hook, pacing and visuals, which is exactly the context a bare LLM prompt lacks.
remix_post grounds the adaptation in the real content: it transcribes the audio and analyzes the scenes first, then produces a consistent title/script/checklist shape every time, joined to the post's stats and outlier score. You skip the media fetch, the transcription, the visual analysis and the prompt engineering, and you pay one flat, refundable credit price instead of stitching four tools together.
20 credits ($0.20) per call in prepaid credits (1 credit = $0.01). For example, 5 calls per dollar. Subscriptions include monthly credits; top-up packs start at $15. Failed asynchronous jobs are refunded automatically, and calls stop at a zero balance, never a surprise bill. See the full pricing table.
Call remix_post with the post URL (or id) and a description of your niche. The post does not need to be tracked: unknown URLs are fetched and ingested automatically. You get back a rewritten hook, a segment-by-segment script, a why-it-went-viral analysis and an execution checklist that keep the proven structure while swapping in your topic and audience.
No. For video posts the remix automatically generates the transcript and visual scene analysis first, then adapts from the full content. You do not have to call request_transcript beforehand; the trade-off is that a video remix takes a little longer than a text-post remix while that enrichment runs.
A remix is asynchronous and usually completes in about one to three minutes, longer for a video post (transcription and visual analysis run first) and for a URL that was not tracked yet (the post is fetched from the platform first). Poll get_remix_result (or get_job_status) with the jobRef to collect it, and the finished idea also appears in the account's Content Ideas dashboard in the web app.
Also available as a REST API endpoint.
Related topics: AI content repurposing · viral format adaptation · content idea generation API · AI script writing from viral posts · niche content brief automation