MCP.Market

Tremelay MCP server

by CharitonMedia·io.github.CharitonMedia/tremelay·v1.0.3·1 stars

Local MCP broker: agents call credentials through a daemon without seeing secrets.

A85/100grade A
What users say
No reviews yet
Be the first
Safety scan
A85/100

not scanned yet

Adoption
New

1 stars

Reviews

Write one

Nobody has reviewed Tremelay yet.

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

Tremelay tools

No tool declarations could be read from the package source. They show once the server is installed.

Not scanned yet. New listings are scanned within a day or two of arriving.

How Tremelay compares

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

Safety score
85/100
higher than 70% of 48,255 servers
GitHub stars
1
more than 65% of 28,688 servers

Release history: 2 versions in the registry since 2026-10-16, latest 2026-10-16, typically one every 1 day.

Install Tremelay in Claude Code, Cursor or VS Code

The registry entry has no remote endpoint or installable package.

What the publisher says

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

Tremelay

A local credential broker, so your AI agents can use your credentials without ever seeing them.

You already give agents real authority: read these issues, push this branch, call that API. Today that means pasting a token into a config file, an environment variable, or an MCP server, and hoping nothing logs it, echoes it, or wanders off to the wrong host with it.

Tremelay splits possession of a secret from authority to use it. You keep credentials in a local encrypted vault. Each agent gets its own identity and its own narrow, expiring, revocable grants — "GET anything under https://api.github.com/repos/you/yourrepo with this token", "authenticate to git.example.com as git with this SSH key". The agent asks Tremelay to perform the operation; Tremelay performs it with the credential and returns only the result the grant allows. The agent never receives the secret, there is no getSecret tool to call, and every attempt — allowed or denied — lands in a hash-chained audit log.

The broker possesses secrets. Agents possess capabilities.

Agents connect over MCP (listcapabilities, describecapability, invoke_capability), or over an ssh-agent socket for SSH signing. Everything is local: one SQLite vault file, one user-only socket or named pipe, no server, no account, no telemetry.

Demo

A real session against a throwaway vault. The secret is the literal string dummy-demo-secret. Every command except vault create also writes one vault_unlock id=… result=allowed line to stderr; those are trimmed here, nothing else is.

$ tremelay vault create --path demo.db
vault_create id=d8df6f8230cf55a9a8b1b7c7df424f1a result=allowed
vault created

$ CRED=$(tremelay credential put --path demo.db --label "demo api key" --type api_key --secret-file ./secret)
credential_put id=649cee93880ae731b17ea858a439d7fe type=api_key result=allowed reasons=
health_assess id=649cee93880ae731b17ea858a439d7fe type=api_key result=allowed reasons=
649cee93880ae731b17ea858a439d7fe

$ AGENT=$(tremelay agent create --path demo.db --label claude-code)
agent_create agent=3a0f2de100565eb2e4be158fdb9a8201 grant= result=allowed
3a0f2de100565eb2e4be158fdb9a8201

$ GRANT=$(tremelay grant create --path demo.db --agent $AGENT --credential $CRED \
    --operation http_broker --resource https://api.github.com/repos/CharitonMedia/Tremelay \
    --method GET --inject bearer --expires 2030-01-01T00:00:00Z)
grant_create agent=3a0f2de100565eb2e4be158fdb9a8201 grant=a0f8b04d64d8ece4a56130feb2607ce9 result=allowed
a0f8b04d64d8ece4a56130feb2607ce9

$ tremelay capability list --path demo.db --agent $AGENT
capability_list agent=3a0f2de100565eb2e4be158fdb9a8201 grant= result=allowed
{"grant_id":"a0f8b04d64d8ece4a56130feb2607ce9","agent_id":"3a0f2de100565

Shortened. The full README is on GitHub.

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

Tremelay: common questions

Is Tremelay MCP server safe?
Yes, by our scan: it is graded A (85/100). Read the Tremelay safety report
Does Tremelay need an API key?
Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is Tremelay maintained?
The last commit was 19 days ago (2026-10-16). The latest release is v1.0.3.
How does Tremelay compare with other MCP servers?
safety score 85/100 (higher than 70% of 48,255 servers); github stars 1 (more than 65% of 28,688 servers). Full comparison.