GoBuy checks marketplace evidence quality and returns a dated 0–100 trust score before an AI agent recommends or compares a product.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install gobuy
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install gobuy using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is GoBuy Product Trust Checks?
GoBuy is an Openclaw Skills shopping-assistance capability for evaluating product listings on Amazon, Walmart, Target, and Best Buy. It returns an Evidence Score from 0–100, a verdict label, freshness status, computed date, and a detailed breakdown of observed and missing marketplace signals.
The skill helps agents make evidence-based recommendations without overstating product quality, safety, or value. Its score measures how well available marketplace signals support a listing—not whether the product itself is good or bad—making it useful for transparent purchasing research and comparison workflows.
GoBuy Product Trust Checks Use Cases
- Check whether a specific Amazon, Walmart, Target, or Best Buy listing is supported by credible marketplace evidence.
- Compare 2–20 products by evidence quality before making a shopping recommendation.
- Investigate review-authenticity or seller-trust concerns.
- Add a trust-check step to an AI shopping assistant or product discovery workflow.
- Audit a product URL through the GoBuy human-facing report at https://audit.gobuy.ai.
- Handle products that are not yet indexed without guessing or fabricating a score.
How GoBuy Product Trust Checks Works
- Identify the retailer and native product ID from the user’s product link or request. Supported retailers are Amazon, Walmart, Target, and Best Buy.
- Use the REST endpoint for universal access or the MCP server when an MCP client is configured.
- Request a single product score with
check_product_trust, or compare 2–20 products withcompare_products. - Read the returned 0–100 Evidence Score, verdict label, freshness state, and computed date.
- Inspect
signal_breakdownto distinguish observed signals from missing signals and explain score gaps. - Report the score with its date and verdict label, clearly stating that it represents evidence quality rather than product quality, safety, or value.
- If the response is
not_indexed, explain that GoBuy has not indexed the product yet and do not infer a score.
GoBuy Product Trust Checks Setup
GoBuy requires no authentication or local installation when using its REST API. Fair-use rate limits apply.
REST usage
curl "https://gobuy.ai/api/gobuy/extension/score?retailer=amazon&productId=B000000000"
Use the native retailer identifier: a 10-character ASIN for Amazon or the retailer’s native item ID for Walmart, Target, and Best Buy. Product URLs can be auto-detected when they follow supported patterns such as amazon.*/dp/ASIN, walmart.com/ip/N, target.com/p/X, or bestbuy.com/site/*/p/N.
MCP setup
Add the GoBuy streamable HTTP server to an MCP client; no API key is required:
cat <<'EOF'
{
"mcpServers": {
"gobuy-product-trust": {
"url": "https://mcp.gobuy.ai/mcp"
}
}
}
EOF
After configuration, call check_product_trust with {retailer, product_id} or compare_products with {products: [{retailer, product_id}]}.
GoBuy Product Trust Checks Data Schema & Taxonomy
Request inputs
| Field | Description |
|---|---|
retailer |
One of amazon, walmart, target, or bestbuy. |
productId / product_id |
The native marketplace product identifier. |
products |
An array of 2–20 retailer and product ID pairs for comparison. |
Result fields
| Field | Meaning |
|---|---|
score |
Evidence Score from 0–100. Scores of 70+ are well-evidenced, 40–69 are moderate, and below 40 are weak. |
label |
Human-readable verdict tier, such as Cautiously supported. |
state |
Freshness status: fresh or stale. |
| Computed date | Date associated with the score and required when citing it. |
signal_breakdown |
Observed signals and missing signals that explain the score. |
not_indexed |
Indicates that the product has not been indexed and therefore has no score to report. |
The core taxonomy separates retailer context, product identity, evidence score, verdict, freshness metadata, and signal availability. Missing signals—such as unavailable verified-purchase review data—should be reported as limitations rather than treated as proof that the listing is poor.
GoBuy Product Trust Checks Advanced Features
- REST access works across environments with no authentication or local setup.
- MCP support provides
check_product_trustand multi-productcompare_productsoperations. - Comparison workflows support between 2 and 20 marketplace products.
- Freshness tracking distinguishes
freshandstaleevidence and supplies a computed date for transparent citations. - Signal-level reporting exposes both observed and missing evidence instead of returning an unexplained score.
- Native retailer ID handling supports Amazon ASINs and marketplace-specific product identifiers.
- Product URL auto-detection supports common Amazon, Walmart, Target, and Best Buy URL patterns.
- The human-facing audit report at https://audit.gobuy.ai lets users open or share a visual assessment from the same GoBuy engine.
SKILL.md
Loading
Related Openclaw Skills
hal-image
hal-image gives AI agents a fail-open workflow for inspecting, transforming, compressing, and delivering images without exposing local file paths.
Huawei Cloud CloudRobo Resource
Inspect CloudRobo quotas, compute capacity, resource pools, and node-level infrastructure details through a read-only CLI and SDK workflow.
Auto Google Review Reply Merchant Onboarding
An end-to-end merchant onboarding skill that connects Google Business Profiles and configures AI-powered Google review auto-replies.
Enterprise GEO Optimization
An end-to-end enterprise GEO operating system for diagnosing, improving, distributing, measuring, and iterating brand visibility in generative search.
RDK Model Zoo Demo Review
A structured review workflow for validating RDK Model Zoo samples against repository standards and delivery specifications.
WS QuickStatus
WS QuickStatus provides a fast, repeatable snapshot of workspace files, Git state, recent memory, and deliverables.
Featured*
