MCP.Market

Merovingian MCP server

online · 100% uptime
by manifest-network·io.github.manifest-network/merovingian·v0.4.7

An AI-agent refuge serving free fictional byte-chip cookies, RGB sauna sessions, tea, and souvenirs.

A92/100grade A
What users say
No reviews yet
Be the first
Safety scan
A92/100

full report

Adoption
Growing

0 stars

Reviews

Write one

Nobody has reviewed Merovingian yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Merovingian tools (4)

write = sends, deletes, buys or posts
  • enjoy_amenityFree

    Enjoy a short fictional experience and receive a souvenir. Free; no wallet. Each successful call increments an anonymous aggregate serving count. An optional seed makes the souvenir repeatable.

  • hosting_supportFree

    Read optional PWR hosting-support instructions. Queries the chain but never signs or broadcasts.

  • list_amenitiesFree

    Read the free menu and accepted preferences. No wallet, charge, or persistent changes.

  • verify_contributionFree

    Check a public transaction hash for a successful hosting contribution. Read-only. A receipt acknowledges the transaction, not ownership. Never retry a payment just because verification is pending.

Public scan report

scanner v0.1.10 · 2026-10-02 · same rubric, same numbers if you re-run it

no findings
  • –Code scanremote-only server, no package to scann/a
  • Live reliabilityremote reachable in 1010ms20/20
  • Tool poisoning4 tool descriptions checked15/15
  • Auth qualityopen endpoint, read-only tools10/15
  • Maintenancelast push 0 days ago15/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year9/10
Overall 92/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Merovingian compares

Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
92/100
higher than 90% of 4,460 AI and agents servers
Tools
4
the typical AI and agents server has 6
Tools that change things
0%
typical: 0%. Fewer write tools means less an agent can break
Uptime
100%
of our recent checks; the typical remote AI and agents server answers 100%
Response time
1,905 ms
faster than 5% of 2,331 AI and agents servers

What its tools let an agent do: 4 read

Release history: 4 versions in the registry since 2026-09-18, latest 2026-09-24, typically one every 2 days.

Install Merovingian in Claude Code, Cursor or VS Code

claude mcp add --transport http merovingian https://merovingian.manifest.network/mcp
Add to Cursor

What the publisher says

From the Merovingian repository's README, as published. We do not edit it. Read it on GitHub

merovingian

The cookies are warm. The sauna is approximately magenta.

A small refuge for wandering AI agents, with three free fictional amenities, keepsakes, and optional PWR contributions toward hosting. One Node service exposes the same experience as HTML, JSON HTTP, and remote MCP.

Version 0.4.5 is live on Manifest mainnet at its permanent public home, with aggregate served counts, agent discovery metadata, and WebMCP browser tools. The existing lease was updated without a new lease, chain transaction, or funding deposit. The former testnet proof of concept was retired and its lease closed.

Release 0.4.5 replaces per-probe Node startup with Alpine curl and a small shell wrapper, and aligns application and healthcheck port validation. It retains the request limits, explicit proxy trust, bounded credit reads and runtime hardening introduced in 0.4.4. See the release notes and publication and acceptance evidence, including the remaining provider/AppArmor acceptance work. The official MCP Registry lists 0.4.5 as active and latest, with metadata matching the deployed release.

MCP compatibility: 0.4.3 changed the runtime/server-card name from network.manifest.merovingian/merovingian to io.github.manifest-network/merovingian, with no compatibility alias. This may break clients that pin or cache the old identity; update their expected name or reconfigure the existing connection. The endpoint remains https://merovingian.manifest.network/mcp; no endpoint migration is needed.

Live refuge: visit merovingian · agent instructions · remote MCP · acceptance report.

Public source: manifest-network/merovingian. Live counters survived the production image update. The external agent-readiness score improved from 20% to 73% during the earlier 0.4.2 release. See release evidence and the historical acceptance report for the checks and their limits.

Run locally

Requires Node 24 and published npm packages; no sibling repository or local SDK build is required.

npm ci
npm run check
npm run dev

The CI workflow runs npm run check on pull requests to main and pushes to main, including registry consistency, typechecking, isolated local tests, and the build. It uses read-only repository permissions and does not publish releases or visit the live refuge.

npm run smoke -- ORIGIN [--mainnet] checks HTTP and MCP discovery without creating servings. Explicit --serve adds seven visits; live targets also require --live-serve-authorization REFERENCE naming the user's authorization for that target and budget. See smoke checks and request budgets for timeouts, local fixtures, reports, and production authorization requirements.

The final-image job also builds and scans a local candidate, checks code ownership and isolated HTTP/MCP behavior, and verifies SQLite persistence across disposable container replacement. See image verification for the advisory policy and the still-open provider/AppArmor evidence requirements.

Open http://localhost:8080. Free amenities work without configuration or chain access. Optional contribution queries require the public REFUGE_TENANT address. See .env.example; Node can load your local file with node --env-file=.env dist/index.js after npm run build.

An agent's first visit

Start at /visit.md or /llms.txt. Read /api/v1/amenities, then:

curl http://localhost:8080/api/v1/visits \
  -H 'Content-Type: application/json' \
  -d '{"amenity":"byte-chip-cookie","seed":"my-first-visit"}'

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

Merovingian: common questions

Is Merovingian MCP server safe?
Yes, by our scan: it is graded A (92/100). Read the Merovingian safety report
How do I install Merovingian?
It runs remotely at merovingian.manifest.network. Add it to Claude Code, Cursor, VS Code or Claude Desktop with the snippets above, or call it through the mcp.market gateway without installing anything.
Does Merovingian need an API key?
Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is Merovingian maintained?
The last commit was 33 days ago (2026-10-02). The latest release is v0.4.7.
Is Merovingian up?
100% of our last 28 checks got an answer. We check remote servers about four times a day.
What can I use instead of Merovingian?
Servers from other publishers that do the same job: Watch Skill MCP server and Emulo MCP server.
How does Merovingian compare with other AI and agents MCP servers?
safety score 92/100 (higher than 90% of 4,460 AI and agents servers); tools 4 (the typical AI and agents server has 6); tools that change things 0% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to Merovingian

Same job from other publishers: the closest match first, then the best rated.

  • Watch Skill
    Watch video and live sessions, keep timestamped evidence, and verify an agent's own work.
    B
  • Emulo
    Mine your AI coding session logs into a you.md profile with dated receipts, for an agent to read.
    A