About VerifiedPricing
VerifiedPricing provides Python REST and CLI clients plus MCP tools for source-linked SaaS pricing data. It supports price lookups, comparisons and explicit requests for pricing coverage, with citations and snapshot details in results.
Key features
- 01Search products and retrieve prices
- 02Compare prices with eligibility reasons
- 03Inspect citations, snapshots and freshness
- 04Submit and track coverage requests
- 05Check account usage and allowance
Use cases
- 01Research a SaaS product’s current pricing
- 02Compare competitor prices and billing conditions
- 03Audit sources and dates behind price records
- 04Request coverage for an unlisted product
MCP support
Provides hosted Streamable HTTP and local stdio MCP options. Its tools expose pricing search, lookups, comparisons, source records, usage and optional coverage requests.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://verifiedpricing.com/mcpclaude mcp add --transport http pucilpet-verifiedpricing https://verifiedpricing.com/mcp{"mcpServers":{"pucilpet-verifiedpricing":{"type":"http","url":"https://verifiedpricing.com/mcp"}}}Open in Cursor{"servers":{"pucilpet-verifiedpricing":{"type":"http","url":"https://verifiedpricing.com/mcp"}}}Open in VS Code{"mcpServers":{"pucilpet-verifiedpricing":{"type":"http","url":"https://verifiedpricing.com/mcp"}}}{"mcpServers":{"pucilpet-verifiedpricing":{"serverUrl":"https://verifiedpricing.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
https://verifiedpricing.com/mcpAuth requiredTools
Details
From the project READMEVerifiedPricing agent clients
Python REST client, CLI and MCP tools for pricing snapshots, source citations, comparisons and explicit research requests.
Bundle v0.3.2 pairs verifiedpricing-client 0.1.3 with verifiedpricing-mcp 0.3.2. The CLI and local stdio support Python 3.10+ on Linux and macOS. Self-hosted HTTP extras require Python 3.11+. Connecting to the hosted endpoint requires no Python installation.
Get started
Create your own key at VerifiedPricing Account.
The account form initially selects prices:read and comparisons:read;
API key creation with scopes omitted defaults to prices:read only. Add the optional
coverage:request scope if you want to submit pricing research requests.
Keep your key in private environment or credential storage; do not put it in
command arguments, URLs, tool arguments or public configuration.
Bundle v0.3.2 uses these wheel filenames:
verifiedpricing_client-0.1.3-py3-none-any.whlverifiedpricing_mcp-0.3.2-py3-none-any.whl
Download both wheels and SHA256SUMS from the release assets, verify their
checksums, then install in an isolated environment:
sha256sum -c SHA256SUMS
python3 -m venv ~/.venvs/verifiedpricing-agent
~/.venvs/verifiedpricing-agent/bin/python -m pip install \
./verifiedpricing_client-0.1.3-py3-none-any.whl \
./verifiedpricing_mcp-0.3.2-py3-none-any.whl
On macOS, use shasum -a 256 -c SHA256SUMS. The REST client uses the Python
standard library; the MCP package installs the pinned official SDK mcp==2.3.0
and its dependencies. Activate the environment before using verifiedpricing
or configuring verifiedpricing-mcp as a local stdio server.
Hosted MCP
Connect a client that supports Streamable HTTP to https://verifiedpricing.com/mcp.
Send your personal read key in Authorization: Bearer YOUR_VERIFIEDPRICING_API_KEY.
Store the real key in your client's private secret or environment settings; never
put it in the URL or a public configuration file. The JSON below shows the shape
used by clients that accept mcpServers HTTP configuration; client support and
secret substitution vary.
{
"mcpServers": {
"verifiedpricing": {
"type": "http",
"url": "https://verifiedpricing.com/mcp",
"headers": {
"Authorization": "Bearer YOUR_VERIFIEDPRICING_API_KEY"
}
}
}
}
The hosted endpoint uses the same personal account allowance. Its MCP Registry entry, metadata version 0.3.2, records the service endpoint. Use the tool list returned by the running service when connecting. The local 0.3.2 package provides the eleven tools below.
Eleven MCP tools
search_products, get_prices, get_price, compare_prices,
get_competitor_prices, get_price_history, get_sources,
get_reviewed_records, get_usage, request_price_coverage,
get_price_coverage_request.
Coverage tools require an API with coverage requests enabled.
request_price_coverage is the optional write tool. It requires
coverage:request, explicit consent: true, and a product ID or canonical
domain. A new domain also needs a supported category. Repeating the same request
returns its original ID; use get_price_coverage_request to follow its status.
Price lookups remain reads when coverage is missing.
REST, CLI and MCP share the same account budget: the free allowance is
1,000 weighted units per day and 10,000 per month. get_usage returns the account's
limits and remaining allowance. A new key or MCP session does not reset them.
Source-backed results
Client 0.1.3 and MCP 0.3.2 retain catalog provenance, amount support, comparison eligibility, conditions, billing and provider retrieval dates in human output. Catalog entries carry the Automated pricing source label.
Responses retain snapshot IDs, source URLs, observation and review dates,
freshness, integrity and comparison eligibility. Automated observations and
recorded human review have distinct labels. Comparisons preserve
not_comparable and its reasons when inputs need more context.
License and data terms
Client and gateway code are MIT licensed. Copyright (c) 2026 VerifiedPricing
contributors. Include the copyright and permission notice from LICENSE with
copies or substantial portions of the software.
The code license does not license service data or cited vendor content. Service terms govern API use and data reuse; retain citations, snapshot IDs, original dates and returned trust fields.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.