# Neoavis (remote · neoavis.com)

Can an AI agent buy on this French store? 500 stores tested: score, guest checkout, UCP. Free.

- Trust score: 77/100 (medium)
- Change this week: +6
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-11-05

## Components

- remote · `neoavis.com`: 77/100 (this document), [markdown](https://verifymcp.io/servers/com-neoavis-constats/api-mcp.md), [page](https://verifymcp.io/servers/com-neoavis-constats/api-mcp)

## Channel facts

- Endpoint: `https://neoavis.com/api/mcp`
- Transports: `streamable-http`
- Auth: `none`
- Version: `1.3.0`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-11-05.

- **Endpoint Security**: 80/100
  - The endpoint's TLS certificate is valid, in date, and uses a strong key.
  - No authorisation is required to call this server. Nothing on it is marked destructive, but we can't show that an anonymous caller can only read: full credit needs the server to list at least one tool and every tool to be verifiably read-only. We use a best-effort check of tool names and descriptions to tell read-only tools from tools that change things, so some genuinely read-only tools may not get full credit. Names that start with a read verb (get, list, read, search and similar) and accurate readOnlyHint annotations help. Meet that to earn full credit, or require authorisation.
  - HTTPS is enforced; there's no plaintext access path.
  - The HSTS (Strict-Transport-Security) header is present.
  - DNSSEC check failed: this domain isn't protected by DNSSEC.
- **Transport & Reachability**: 100/100
  - Verified streamable-http transport via a live MCP handshake.
- **Schema Quality & AI Usability**: 63/100
  - AI-judged instruction clarity (excellent).
  - Context-footprint check failed: tool/resource definitions use about 1550 tokens (~310/item across 5 items; 5 tools + 0 resources), over budget; trim descriptions and params.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 40/100
  - Stability observed for 12 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 100/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 100% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 5 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the Neoavis MCP server?

Neoavis is a hosted endpoint at https://neoavis.com/api/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add --transport http com-neoavis-constats 'https://neoavis.com/api/mcp'
```

### Cursor

```json
{
  "mcpServers": {
    "com-neoavis-constats": {
      "url": "https://neoavis.com/api/mcp"
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "com-neoavis-constats": {
      "type": "http",
      "url": "https://neoavis.com/api/mcp"
    }
  }
}
```

### Codex

```toml
[mcp_servers.com-neoavis-constats]
url = "https://neoavis.com/api/mcp"
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "com-neoavis-constats": {
      "type": "remote",
      "url": "https://neoavis.com/api/mcp",
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add com-neoavis-constats --url 'https://neoavis.com/api/mcp' --transport streamable-http
```

### Hermes

```yaml
mcp_servers:
  com-neoavis-constats:
    url: "https://neoavis.com/api/mcp"
```

### Netclaw

```json
{
  "McpServers": {
    "com-neoavis-constats": {
      "Transport": "http",
      "Url": "https://neoavis.com/api/mcp"
    }
  }
}
```

### Vellum

```bash
assistant mcp add com-neoavis-constats -t streamable-http -u 'https://neoavis.com/api/mcp'
```

### Other

```json
{
  "mcpServers": {
    "com-neoavis-constats": {
      "type": "http",
      "url": "https://neoavis.com/api/mcp"
    }
  }
}
```

The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs.

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-11-04 (score 77, +1)

No change was recorded against any check on this day. Stability & Change Management went from 33 to 37. That category is still filling its 30-day observation window: 10 days of observed history at the previous scan, 11 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-11-02 (score 76, +4)

- [security] The server rewrote its instructions, which are the text every model session reads
- [security] Tool “classement_secteur” rewrote its description, which is the text the model reads
- [security] Tool “comparer_marchands” rewrote its description, which is the text the model reads
- [security] Tool “rechercher_marchand” rewrote its description, which is the text the model reads
- [security] Tool “consulter_marchand” rewrote its description, which is the text the model reads
- [security] Tool “trouver_marchands” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: 136 → 310
- [functional regression] Schema quality: 136 → 170
- [functional improvement] Tool coverage: 69% → 100%
- [functional improvement] MCP protocol: fail → pass
- [functional] Schema quality: good → excellent
- [functional] MCP protocol version: 2025-06-18 → 2025-11-25
- [functional] Server version: 1.2.0 → 1.3.0
- [functional] Server version: 1.1.0 → 1.2.0
- [cosmetic] “trouver_marchands” added an optional parameter “limite”
- [cosmetic] “classement_secteur” reworded the description of “langue”
- [cosmetic] “classement_secteur” reworded the description of “secteur”
- [cosmetic] “comparer_marchands” reworded the description of “domaines”
- [cosmetic] “comparer_marchands” reworded the description of “langue”
- [cosmetic] “consulter_marchand” reworded the description of “domaine”
- [cosmetic] “consulter_marchand” reworded the description of “langue”
- [cosmetic] “rechercher_marchand” reworded the description of “langue”
- [cosmetic] “rechercher_marchand” reworded the description of “requete”
- [cosmetic] “trouver_marchands” reworded the description of “frais_livraison”
- [cosmetic] “trouver_marchands” reworded the description of “langue”
- [cosmetic] “trouver_marchands” reworded the description of “note_min”
- [cosmetic] “trouver_marchands” reworded the description of “prix_livre”
- [cosmetic] “trouver_marchands” reworded the description of “retours”
- [cosmetic] “trouver_marchands” reworded the description of “sans_compte”
- [cosmetic] “trouver_marchands” reworded the description of “secteur”
- [cosmetic] “trouver_marchands” reworded the description of “ucp”

### 2026-10-31 (score 72, +1)

No change was recorded against any check on this day. Stability & Change Management went from 20 to 23. That category is still filling its 30-day observation window: 6 days of observed history at the previous scan, 7 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-29 (score 71, +3)

- [security improvement] HSTS header: fail → pass

### 2026-10-27 (score 68, +1)

No change was recorded against any check on this day. Stability & Change Management went from 7 to 10. That category is still filling its 30-day observation window: 2 days of observed history at the previous scan, 3 at this one. The score rises as the window fills, whether or not the server changes.

### 2026-10-26 (score 67, 0)

- [security] Tool “consulter_marchand” rewrote its description, which is the text the model reads

### 2026-10-25 (score 67, +1)

- [security] Tool “consulter_marchand” rewrote its description, which is the text the model reads
- [functional regression] Schema quality: pass → fail
- [functional improvement] Tool coverage: 50% → 69%
- [functional improvement] Stability: unverified → 0.03
- [functional] Server version: 1.0.0 → 1.1.0
- [functional] New tool “comparer_marchands”
- [functional] New tool “trouver_marchands”

### 2026-10-24 (score 66)

First indexed and scored.

## MCP tools (5)

### `consulter_marchand` (~340 tokens)

Consulter un marchand

Full test result for ONE French online store, by domain: what an AI shopping agent managed to do there at the latest test. Use it before visiting, recommending or buying from a French store, or when the user asks if a store works with AI agents. To find stores you do not know yet, use trouver_marchands; to look up a store by name, rechercher_marchand; to put several side by side, comparer_marchands. Returns: note (score out of 100: journey 55, proven purchase facts 30, machine readability 15), etapes (six steps in order — acces, cookies, recherche, fiche, panier, caisse — each reussi, partiel, echec or non_atteint, with a motif when it failed), faits (oui, non or inconnu: checkout without an account, delivered price before sign-up, shipping costs, return policy, third-party seller), machine (product data, robots.txt, llms.txt, sitemap, UCP), ucp_commande (store declares UCP checkout), dernier_test (date), test_ancien (true if older than 90 days), resume (one-sentence summary to quote) and page (link to cite). Unknown or untracked domain: error result, nothing charged.

Input parameters:

- `domaine` (string, required): Store domain, for example respire.co. A full URL or a www. prefix is accepted and normalised.
- `langue` (string): Language of the page links and, for consulter_marchand, of the summary sentence: fr (default) or en. Data fields are the same in both.

### `classement_secteur` (~230 tokens)

Classement d'un secteur

Ranking of ALL tested French online stores in one sector, best score first. Use it when the user wants the best stores of a sector as a whole (for example the best beauty stores for an AI agent). To filter by proven facts such as guest checkout, use trouver_marchands; for the details of one store, consulter_marchand. Returns: secteur and marchands, a list of every ranked store (30 to 70 per sector) with rang, ex_aequo (equal scores share a rank, then alphabetical order), domaine, nom, note out of 100, nombre_avis, dernier_test, test_ancien and page. Sectors: beaute, mode, maison, sport, epicerie, enfant, animalerie, bijoux, bricolage, high-tech. Unknown sector: error result listing the valid ones.

Input parameters:

- `langue` (string): Language of the page links and, for consulter_marchand, of the summary sentence: fr (default) or en. Data fields are the same in both.
- `secteur` (string, required): Sector id.

### `rechercher_marchand` (~208 tokens)

Rechercher un marchand

Find a tracked French store by part of its NAME or DOMAIN, across all sectors (case and accents ignored). Use it when the user names a store but you do not know its exact domain; then call consulter_marchand with the domain found. To discover stores by criteria rather than by name, use trouver_marchands. Returns: requete and marchands, every match (usually a few) with rang in its sector, ex_aequo, secteur, domaine, nom, note out of 100, nombre_avis, dernier_test, test_ancien and page. No match: an empty list, which means the store is not tracked yet.

Input parameters:

- `langue` (string): Language of the page links and, for consulter_marchand, of the summary sentence: fr (default) or en. Data fields are the same in both.
- `requete` (string, required): At least two characters of the store name or domain, for example nocibe or sezane.

### `trouver_marchands` (~393 tokens)

Trouver des marchands par critères

Discover French online stores where an AI agent can actually buy, filtered by PROVEN facts, in one sector or all of them. Use it when the user asks where to buy something online in France, which stores allow guest checkout, or which show the delivered price. To look up a store by name, use rechercher_marchand; for a whole sector ranking, classement_secteur. Every filter is optional; set a boolean to true to require that fact. Returns: secteur, criteres (filters applied), total (number of matching stores) and marchands, best score first, at most limite entries, each with rang, ex_aequo, secteur, domaine, nom, note out of 100, the five facts as booleans (sans_compte, prix_livre, frais_livraison, retours, ucp_commande), dernier_test and page. No match: total 0 and an empty list.

Input parameters:

- `frais_livraison` (boolean): Shipping costs were found.
- `langue` (string): Language of the page links and, for consulter_marchand, of the summary sentence: fr (default) or en. Data fields are the same in both.
- `limite` (integer): Maximum number of stores returned (default 20, at most 50); total still counts every match.
- `note_min` (number): Minimum score out of 100.
- `prix_livre` (boolean): Total price including delivery was shown before any sign-up.
- `retours` (boolean): Return policy was found and quoted.
- `sans_compte` (boolean): Agent reached checkout without creating an account (proven by a screenshot or quote).
- `secteur` (string): Restrict to one sector; all sectors if omitted.
- `ucp` (boolean): Store publishes a UCP profile that declares checkout, so an agent can order through the protocol.

### `comparer_marchands` (~190 tokens)

Comparer des marchands

Side-by-side test results for two to five French online stores you already know, in the order given. Use it when the user hesitates between specific stores. To find candidates first, use trouver_marchands or rechercher_marchand. Returns: marchands, one full result per store, identical to consulter_marchand (score out of 100, six steps acces to caisse, proven facts, machine readability, UCP checkout, summary and page). Each store counts as one request against the rate limit. Fewer than two or more than five domains: error result.

Input parameters:

- `domaines` (array, required): Two to five store domains, for example ["respire.co", "typology.com"].
- `langue` (string): Language of the page links and, for consulter_marchand, of the summary sentence: fr (default) or en. Data fields are the same in both.

## Diagnostics

Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/com-neoavis-constats/api-mcp#diagnostics

## Score history

- 2026-11-05: 77
- 2026-11-04: 77
- 2026-11-03: 76
- 2026-11-02: 76
- 2026-11-01: 72
- 2026-10-31: 72
- 2026-10-30: 71
- 2026-10-29: 71
- 2026-10-28: 68
- 2026-10-27: 68
- 2026-10-26: 67
- 2026-10-25: 67
- 2026-10-24: 66

## Common questions

### What is the Neoavis MCP server?

Neoavis is an MCP server listed in the public MCP registry as com.neoavis/constats. Can an AI agent buy on this French store? 500 stores tested: score, guest checkout, UCP. Free. This page covers its hosted endpoint (https://neoavis.com/api/mcp).

### Is the Neoavis MCP server safe to use?

Neoavis scores 77 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Neoavis MCP server expose?

Neoavis exposes 5 tools: consulter_marchand, classement_secteur, rechercher_marchand, trouver_marchands, comparer_marchands. Their descriptions and schemas cost roughly 1,361 tokens of context every time the server is loaded.

### Does the Neoavis MCP server require authentication?

No. We connected to Neoavis without credentials and it answered, so anything it exposes is reachable by anyone who knows the address.

### Is the Neoavis MCP server still maintained?

Neoavis is still listed as active in the MCP registry. We last reached this channel on 5 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What did VerifyMCP test on the Neoavis MCP server?

VerifyMCP ran 13 checks on Neoavis, covering TLS certificate, Authorization, HTTPS, HSTS header and 9 more. Supply Chain Security and Provenance & Transparency do not apply to a hosted endpoint, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.

## Links

- Remote endpoint: https://neoavis.com/api/mcp
- Website: https://neoavis.com/en/agents
- Changelog RSS feed: https://verifymcp.io/servers/com-neoavis-constats/api-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/com-neoavis-constats/api-mcp/changelog.json
- Trust report (JSON): https://verifymcp.io/servers/com-neoavis-constats/api-mcp.json
- HTML version of this page: https://verifymcp.io/servers/com-neoavis-constats/api-mcp