TikTok research
Profiles, videos, hashtags, comments, and public creator signals through provider routing.
Give agents research packets they can inspect before they act.
Data for Agents packages sources, freshness, spend limits, provider state, and the next safe action into one receipt your agent can cite.
Demo packet output
Creator research packet: dry-run provider receipt, 2 source receipts, 18 estimated credits, agent decision = needs_live_run.
curl https://data-for-agents.vercel.app/api/demo-packet
Proof note: see sources, spend cap, provider status, and safe next action first.
Alpha question
What production agent workflow needs external data this week?
Start with one packet template. We will show the source receipts, provider status, fallback notes, and spend caps before live use.
Workflow in, packet plan out
A production agent sends the job it needs to complete. The API returns the route, sources, spend cap, and decision boundary before any live provider charge.
Input workflow
Validate a shortlist of TikTok creators before an outreach agent writes first-touch messages.
Packet plan returned
- Template: creator_research_packet_v1
- Sources: public profile, recent posts, audience cues
- Provider route: TikTok, Instagram, web crawl
- Spend cap: 50 credits, 18 estimated credits
- Agent decision: needs_live_run
Use this packet for
Creator shortlist validation
Check public audience cues, source receipts, and confidence before an outreach agent acts.
Competitor content evidence
Bundle visible claims, page snapshots, and evidence links into one inspectable packet.
Inbox and action evidence
Give agents the receipt trail they need before drafting replies, routes, or follow-up tasks.
Packet templates by outcome
Creator research packet
Audience fit, content hooks, public source receipts, and outreach-ready evidence.
Web evidence packet
Crawl results, screenshots, citations, and confidence notes for agent decisions.
Social proof packet
Public mentions, posts, creator signals, and exportable receipts for campaign workflows.
What a packet can include
YouTube intelligence
Channels, videos, transcripts, search results, and content evidence for agents.
X and Instagram
Public posts, profiles, reels, and social proof without custom scraping glue.
Screenshots
URL, HTML, PDF, and visual evidence capture with provenance metadata.
Web extraction
Clean markdown, crawls, search, and structured outputs for LLM workflows.
Agent inbox
A future unified inbox for Gmail, Slack, Crisp, webhooks, and approvals.
Built around governed runs.
Each packet shows the input, quote, sources, provider status, fallback notes, normalized result, and export path. Visible provider status and fallback notes keep agents from treating missing or partial data as complete. The dry-run packet API now turns a workflow description into a route plan, spend cap, source checklist, and agent decision boundary before live provider charges.
POST /api/packets/run
Content-Type: application/json
{
"workflow": "Validate a shortlist of TikTok creators before an outreach agent writes first-touch messages.",
"requester": "operator@example.com",
"maxCredits": 50
}Inspect before acting
The packet inspection API gives agents a simple usable, needs_live_run, or do_not_use decision with checks for source receipts, provider status, spend boundary, and agent safety.
POST /api/packets/inspect
Content-Type: application/json
{
"packet": { "id": "packet_or_demo_payload" },
"requester": "operator@example.com"
}Credit model preview
Pricing is designed to cap spend before an agent runs external-data work. Start with controlled provider credits each month, then top up only when a workflow needs more live runs.
Free
$0/mo
100 included credits
Starter
$29/mo
2,000 included credits
Pro
$99/mo
10,000 included credits
Scale
$499/mo
60,000 included credits