
Comparr MCP server
Which MCP servers are Comparr's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.
read 2026-10-26
- Registry namefr.comparr/comparrverified in the MCP registry
The registry name sits under comparr.fr. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.
- Stateanswerschecked 2026-10-26
- Tools4
- Server calls itselfcomparr
- Server version1.1.1
- Protocol version2025-06-18
On 2026-10-26 the server answered initialize and tools/list without a sign-in. First answer in 229 ms.
- Its tools do every use case we checked: check whether a discount is a genuine price drop; find the lowest recent price for a product.
Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.
- INITE MCP score100 of 100answers 100 · does the job 100 · definitions 100 · context 100 · connects 100
- What an agent can do with it2 of 2 use cases
- Answered100% of days10 of 10 daily checks
- Median first answer275 ms
- Space in an agent's context≈ 642 tokensthe definitions of all its tools (4), which an agent carries on every turn
- Tool definitionsA · 100 of 100
- Typed results0 of 4 toolsdeclared outputSchema: results come typed, read once
- Every tool passes all six checks.
What an agent can do with it
- can do · check whether a discount is a genuine price drop
- not sure · compare two products criterion by criterion
- can do · find the lowest recent price for a product
Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?
Worth a look
None of the five warning signs in these definitions.
A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →
search_productsCompare the prices of a product category or a specific model across merchants (France on comparr.fr, United States on comparr.co). Returns ranked products with every offer, the exact seller, the item condition and the date each price was…
get_productAll current offers for one product (id or slug from search_products or get_deals), with the date each price was checked, the exact seller, the condition, and a 'is now a good time to buy' verdict computed from Comparr's recorded price…
get_price_historyRecorded price history of one product: best price per day across all offers AND the series of each individual offer, which tells a real price drop apart from a cheaper seller appearing. Readings are periodic, dated snapshots (often…
get_dealsCurrent verified price drops: the SAME offer is cheaper than its own lowest recorded price over the previous 90 days, backed by several days of readings, checked within 48 h, new condition only, Amazon offers re-verified. Each deal…
How to connect
Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:
{
"mcpServers": {
"comparr": {
"type": "http",
"url": "https://www.comparr.fr/mcp"
}
}
}Add it
Claude Code, in a terminal
claude mcp add --transport http comparr https://www.comparr.fr/mcpclaude.ai, ChatGPT and other clients: add https://www.comparr.fr/mcp as a custom connector in the settings.
Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.
Before we count this visit
Analytics and ad measurement tell us which pages and which of our ads do their job: Google Analytics and Google Ads, the Meta pixel, Yandex Metrica, and one first-party id of our own. Nothing is stored until you answer, and declining changes nothing about the site.