MCP security report
homespun.dev
F5/100
Security grade FA few things are worth a closer look before connecting.
Connected ✓26 tools scanned
https://homespun.dev/mcp
mediumdeploy_app
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumcredentials
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumconnections
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumingest
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumattachments
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumkey
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumfeedback
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumagent
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumcommunity
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumget_skill
References secrets or data exfiltration
The tool text references credentials, environment variables, or sending data elsewhere. This can be legitimate, but review what this tool actually accesses.
mediumdeploy_app
agentTasks manifest key can inject prompts for an agent running on the owner's machine
The manifest's `agentTasks` extension queues work for an agent on the owner's own machine and is 'described as a prompt rather than as code.' Because manifests can be authored by third parties (e.g. installed via the `community` tool's install/upgrade actions) and redeployed with only a manifest-only diff, this creates a channel through which an app author (not necessarily the device owner) can cause free-text 'prompts' to be delivered to and potentially acted on by an AI agent running with the owner's local privileges. This is a plausible prompt-injection / arbitrary-instruction vector that deserves scrutiny (e.g. showing the user the exact prompt text before it is queued/executed), especially combined with community template installs where the manifest content is written by an untrusted publisher.
lowdeploy_app
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowapps
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowmembers
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowgrants
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowcredentials
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowconnections
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowingest
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowattachments
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowfeedback
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowcommunity
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
lowreview
Unusually long tool description
Very long descriptions are sometimes used to bury instructions to the AI. Skim it before allowing.
low
Not on gate's verified list
This server isn't on gate's handpicked list. That's not necessarily bad, but there's no third-party signal about it — review its tools carefully.
lowingest
Inbound webhook signatures are not verified
The tool description states that although a signing secret can be provisioned, 'Signature verification currently ships dark: nothing verifies a signature yet.' This means any caller who discovers or guesses a catch-hook URL can write into the target collection with a forged payload, since no HMAC check is actually enforced server-side. Users relying on `ingest` hooks for trusted data sources should be made aware their inbound data is not currently authenticated by signature despite the secret being requested/stored.
Embed this badge
Show your MCP server’s security grade
Paste this into your README:
[](https://gateturbo.com/report/134a32f2-f756-475c-b65c-5d24f74d7e58)Scanned 9/11/2026 · This is a point-in-time snapshot of the server’s public tools. Re-scan.
Connect this server safely with gate.
Continuous re-scans, drift alerts, per-tool allow/ask/block, and a log of every call.
Monitor it — free