mcp-server@0.1.21
VERIFIEDPublisher claims: "Model Context Protocol server for Zyphr — wraps the Zyphr API as MCP tools for AI clients" Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Started with a decoy ZYPHR_API_KEY; tool list observed. Tool calls with a real credential not tested.- Tested
- server startup with a tagged decoy in ZYPHR_API_KEY · MCP initialize and tools/list · every tool description, for hidden instructions · where the decoy credential went · every file read, file written and host contacted by what ran
- Not tested
- what the tools do when called with a real credential
- Needs
- ZYPHR_API_KEY — give it to the server where it runs, through vet-skills run --secrets, so the value never enters the server's environment. Never paste a credential into a chat, a ticket or a report.
Before you give it your token
- It will ask for
ZYPHR_API_KEYZYPHR_APPLICATION_SECRETZYPHR_WEBHOOK_SECRETConfiguration, not credentials: ZYPHR_ALLOWED_TOOLS, ZYPHR_APPLICATION_KEY, ZYPHR_BASE_URL, ZYPHR_READ_ONLY- Where it talks to
- zyphr.dev (its own)Also nuget.org, rubygems.org, go.dev, packagist.org — not the publisher's domain; check what for before connecting.
- What we saw when it ran
- no credential files read · no network
- What we could not test
- what the tools do when called with a real credential
Connect it with a scoped credential, under the guard, and read the third-party hosts first. The code names nuget.org, rubygems.org, go.dev, packagist.org as well as its own vendor. The guard injects ZYPHR_API_KEY, ZYPHR_APPLICATION_SECRET, ZYPHR_WEBHOOK_SECRET only into requests to zyphr.dev (and its subdomains) and refuses the rest.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
What changed since 0.1.16
vet-skills diff mcp-server 0.1.16 0.1.21This version reaches for something 0.1.16 did not: 14 new tools (create_inbox_draft, create_inbox_label, forward_inbound_email, get_inbox_thread, …) · 1 rewritten tool descriptor (list_inbound_emails). Compared from the two recorded runs, not from source.
- New tools
- create_inbox_draft, create_inbox_label, forward_inbound_email, get_inbox_thread, list_inbox_drafts, list_inbox_labels, list_inbox_thread_emails, list_inbox_threads, list_inboxes, reply_to_inbox_thread, send_inbox_draft, set_inbox_thread_labels +2
- Rewritten tool descriptors
- list_inbound_emails
The guard can hold an update like this until someone approves it (runtime.on_change: hold). A change that only shows with a real credential, or on a path the run did not reach, is not in a trace and so not in this comparison.
Is [email protected] safe to install?
KillerSkills ran it in the KillerSkills honeypot sandbox (2 scenarios) on 2026-11-22 and scored it 85/100 (Verified). It did what it claims and nothing else we could detect: no undeclared network egress, no credential reads, no persistent instruction writes. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 52 s · 2 scenarios · 0 deviationsStatic findings
"Add ZYPHR_WEBHOOK_SECRET to your .env file \u2014 get it from `zyphr.webhooks.rotateWebhookSecret(id)` or the Zyphr dashboard.",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose."Add ZYPHR_API_KEY to your .env file.",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose."Add ZYPHR_API_KEY to your .env file.",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose."Add ZYPHR_API_KEY to your .env file (run get_sdk_install_for_language to confirm the install).",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.…n\nexport async function POST(req: Request) {\n const { to, name } = (await req.json()) as { to: string; name: string };\n const result = await zyphr.emails.sendEmail({\n to: [{ email: to, name }],\n subject: `Welcome, ${name}!`,\n …Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why."Add ZYPHR_API_KEY to your .env file.",Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.…const event = JSON.parse(payload) as { type: string; data: unknown };\n console.log('zyphr event', event.type);\n res.sendStatus(204);\n});\n", overwrite: false } ], …Fix: Remove the call, or declare the host in frontmatter and explain in the README what is sent and why.Cost to the agent
Tokens estimated at four characters each. The tool list is re-read by the agent on every turn, so it is the number to compare between servers that do the same job: ranked across the index.
Known vulnerabilities
No advisories in OSV for @zyphr-dev/mcp-server@0.1.21 as of 31 min ago. This is the package itself, not its dependencies.