3 retailers verifiedConfidence 0.89Grade B evidence
notifications_active
Slack alerts
Threshold breaches reach the right channel the moment a run finds them.
upload_file
CSV import
Bring your catalog in bulk and edit it with bulk operations.
terminal
REST API & MCP server
Build on it: a programmatic REST API and a Model Context Protocol server let external agents, like Claude, operate your workspace.
sync_alt
Guardrailed write-back
Push approved prices to Shopify, WooCommerce, BigCommerce, and Salesforce Commerce Cloud, or self-hosted stores via a signed webhook.
bolt
Autonomous pricing
Opt in to guardrailed auto-apply on Max and Enterprise — approved price moves ship without a manual click.
how_to_reg
Approval workflow
An approver role reviews every write-back, with approve/reject and a 7-day rollback if a move needs undoing.
fact_check
Quality grading
A rubric agent scores the evidence so weak data never drives a decision.
tune
Alert thresholds
Set breach thresholds per product or per tag so alerts fire exactly where you need them.
A plan for every stage
Annual billing, no password, cancel anytime. See current prices on the signup page.
target
You don't have to watch every SKU — just the ones that move your margin
Track the hero SKUs that actually drive your margin. Tags, threshold overrides, and refresh cadence cover the long tail automatically. A catalog bigger than a single run's shape isn't a blocker — it's covered by running more tag-scoped batches (not a bigger plan).
Plan
Tracked catalog
Products per check
Competitors tracked
Basic
100
5
3
Pro
350
25
10
Max
1,500
60
12
Basic
For solo operators getting started.
checkTrack 100 products vs 3 top competitors
check5 products per check, rotating through your catalog
checkAI quality-graded checks & recommendations
checkNo credit card · business email
check30 days to prove the ROI
infoStarter access, no overreach — sign up with a personal email and unlock full access with a credits purchase.
A dedicated pricing analyst runs roughly $80,000 a year (BRD §4.1). This is a cost comparison, not a savings guarantee — your results depend on your own catalog and market.
Estimate only — your actual charge is confirmed at checkout.
Prices in USD — you can pay in your local currency at checkout.
lock
Hard caps by default: a run past your plan's included allocation is refused, never billed.
toggle_off
Usage-based overage is opt-in per workspace, and only available on Pro and Max today — you decide if and when it turns on.
Starting at $0.05 per research unit.
sell
Prepaid research credits are a one-time purchase on Basic, Pro, and Max. They never auto-renew.
Coming soon
What’s next on the roadmap — not yet shipped, so not yet gated by a plan.
Coming soonhub
Bring your own cloud
Run the agents on your own OpenAI, Bedrock, or Vertex credentials — on top of the managed Anthropic, Bedrock, and Vertex AI runtimes already built in.
Coming sooninventory_2
Procurement intelligence
A workflow pack that turns the same agent pipeline toward supplier and procurement pricing.
Coming soonapps
Vertical packs
Workflow packs that point the same pipeline at new spend categories — starting with marketing-spend monitoring across Google, Meta, and LinkedIn Ads.
Coming soonauto_fix_high
Autonomous onboarding agent
An agent that sets up your catalog and retailer list for you from a short conversation.
Review recommendations, approve price moves, and catch alerts — wherever you are. Join the waitlist.
Frequently asked questions
What is the Agnotiq MarginTide Price Checker Agent?expand_more
It is a multi-tenant SaaS for competitive pricing intelligence. A managed agent pipeline checks competitor prices on a schedule or on demand, grades the evidence for quality, and surfaces margin-aware recommendations in a dashboard.
How does the pricing-intelligence agent work?expand_more
A Composer agent plans each run, parallel Researcher agents gather live competitor prices from the open web, and a Quality Grader scores the evidence against a rubric before any recommendation reaches you.
Which platforms can it write prices back to?expand_more
Approval-gated, guardrailed price write-back is available for Shopify, WooCommerce, BigCommerce, and Salesforce Commerce Cloud, plus a signed outbound webhook for self-hosted stores. Basic is read-only.
Can external tools or agents operate my workspace?expand_more
Yes. A programmatic REST API and a Model Context Protocol (MCP) server let external tools — and AI agents such as Claude — trigger runs and read results, under the same tier gates and audit trail as the dashboard.
Which AI runtimes can the agents run on?expand_more
The pipeline runs on Anthropic Claude via managed agents today. The same pipeline can also run on AWS Bedrock and Google Vertex AI managed runtimes, currently in beta. Bringing your own provider credentials (including OpenAI) is on the roadmap.
Do I need a password or shared secret to sign in?expand_more
No. Sign-in is passwordless. You receive a secure magic link or a six-digit code by email. There is no password to set or reset.
What plans are available?expand_more
Four tiers: Basic, Pro, Max, and Enterprise. Pricing is shown on the signup page and billed through Stripe; you can cancel anytime from the billing portal.
What is an AI price check (research unit)?expand_more
1 product × 1 competitor × 1 refresh = 1 AI price check (one research unit).
Is my workspace data isolated from other tenants?expand_more
Yes. Every workspace-scoped table is protected by Postgres row-level security, with an additional application-layer workspace filter on top.
Stop guessing what your competitors charge
Set up your catalog, pick a plan, and let the agents do the watching.
