Starts monitoring a profile so it is automatically re-crawled at a cadence you choose (daily, every_3_days or weekly), keeping its posts and stats fresh without you polling crawl_profile. Identify the profile by profileId, platform+handle, or a public profile/post URL. Managing monitoring is free; each scheduled refresh crawl costs credits (10 per refresh) and monitoring pauses itself if your balance runs out, then resumes when you top up. The profile must already be in the database; crawl_profile it first if it is not. Calling again on an already-monitored profile just updates the cadence. Pull the new posts with get_tracked_updates.
Put a creator profile on autopilot: track_profile re-crawls it automatically at a cadence you choose (daily, every 3 days or weekly), keeping its posts and stats fresh without you polling crawl_profile by hand. Identify the profile by id, platform+handle, or a public URL. Managing monitoring is free; you pay only for the scheduled refresh crawls (10 credits each), and monitoring pauses itself if the balance runs out.
Connect the Viral Outliers MCP server once and the track_profile 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 track_profile on its own whenever the task needs it.
| Name | Type | Description |
|---|---|---|
profileId | string | Internal profile id from search results. |
platform | string | "tiktok" | "instagram" | "youtube" (pair with handle). |
handle | string | Public handle, without @ (pair with platform). |
url | string | Public profile or post URL identifying the profile. |
frequency | string | "daily" | "every_3_days" | "weekly" (default weekly). |
The call confirms monitoring is active for the profile at the cadence you set. From then on the profile is re-crawled on that schedule automatically, and each refresh pulls in whatever the account has posted since the last check. You read those new posts with get_tracked_updates, and search, get_profile and the outlier scores all reflect the freshened data.
Starting, changing and stopping monitoring are free. The cost is the recurring refresh crawls: 10 credits per scheduled refresh, so weekly monitoring is roughly 40 credits a month and daily roughly 300. If the balance cannot cover a refresh, monitoring pauses itself and resumes automatically once you top up, so it can never overspend.
You identify the profile by profileId, by platform+handle, or by a public profile/post URL, and set a frequency (default weekly). The profile must already be in the database; if it is not, run crawl_profile first, then track it. Calling track_profile again on an already-monitored profile just updates the cadence or re-activates a paused one, it never creates a duplicate or forces an extra paid crawl.
Each scheduled refresh is a posts_crawl (cheaper than a first full profile crawl) charged against your credit balance at refresh time. The refresh only bills for the new work: pagination plus thumbnails for genuinely new posts. Because pausing on an empty balance is automatic and reversible, a monitored set degrades gracefully rather than erroring when credits run low.
Competitor watch is the headline: track a rival account weekly, call get_tracked_updates on a schedule, and alert the user whenever a new post lands, optionally kicking off remix_post on it. Fresh-dashboard workflows track a client's whole set of creators daily so search and get_profile always reflect current stats. Trend desks monitor the accounts driving a niche so their outlier scores stay current.
The pattern replaces a brittle "re-crawl everything on a cron" job with managed, per-profile scheduling that bills only for real refreshes and pauses itself when it should.
Rolling your own monitoring means running a scheduler, deciding per profile when it is due, calling crawl_profile on a timer, deduping what comes back, and building your own guardrail so a drained balance does not either overspend or silently stop. Every one of those is a place to leak money or miss posts.
track_profile makes the schedule a managed property of the profile: you set a cadence once, refreshes run and bill themselves at the cheaper posts-crawl rate, new posts surface through get_tracked_updates as a clean incremental feed, and low-balance handling (pause and auto-resume) is built in. You get monitoring as a feature instead of infrastructure to maintain.
This skill is free and never spends credits.
Call track_profile with the profile and a cadence (daily, every 3 days or weekly). The profile is then re-crawled on that schedule automatically, and get_tracked_updates returns whatever new posts each refresh brought in, with no manual re-crawling or polling loops.
Starting, changing or stopping monitoring is free. You only pay for the scheduled refresh crawls: 10 credits each, so weekly is roughly 40 credits a month and daily roughly 300. Monitoring pauses automatically when your balance runs out and resumes after you top up, so it can never overspend.
It simply updates the refresh cadence and re-activates monitoring; it never creates a duplicate or forces an extra paid crawl. Re-tracking is the way to change a profile from weekly to daily or to resume a paused one.
Monitoring pauses itself when a scheduled refresh cannot be charged, so it never overspends or fails loudly. It resumes automatically the next cycle after you top up, with no need to re-track. You can see which profiles are paused, and why, with list_tracked_profiles.
Also available as a REST API endpoint.
Related topics: social media monitoring API · scheduled profile crawl · creator tracking automation · competitor alerting · automatic profile refresh · creator change monitoring