unifi-fabric-mcp-server@0.7.2
CAUTION· not startedPublisher claims: "MCP server for UniFi Fabric: Network, Protect, Site Manager, InnerSpace, Carrier Fabric, Mobility." Observed in sandbox: no network, touches nothing outside the project, writes no memory.
Coverage
Not started: dependency install failed; the server itself was not observed.- Tested
- the dependency install step · every file read, file written and host contacted by what ran
- Not tested
- server startup · the tool list · tool behavior
What this means for the label: Caution here covers what ran. The server's tools were not exercised with a real credential, so nothing about their behavior is asserted; the credential summary below says what to do with that.
Before you give it your token
- It will ask for
- no credential variables read in the code
- Where it talks to
- no vendor host named in the codeAlso ui.com, openapis.org, githubusercontent.com — 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
- server startup · the tool list · tool behavior
Connect it under the guard. Not started: dependency install failed; the server itself was not observed. The code names no credential variable; the guard pins the network to what the honeypot saw and logs every call.
Never paste a credential into a chat, a ticket or a report, including ours. How the guard holds it.
What changed since 0.7.0
vet-skills diff unifi-fabric-mcp-server 0.7.0 0.7.2This version reaches for something 0.7.0 did not: 3 new hosts (githubusercontent.com, openapis.org, ui.com). Compared from the two recorded runs, not from source.
- New hosts
- githubusercontent.com, openapis.org, ui.com
0.7.0 was not run in the sandbox (install failed); the comparison covers what static analysis recorded only.
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; not started: dependency install failed; the server itself was not observed) on 2026-11-20 and scored it 69/100 (Caution). It is not known to be malicious, but it has behaviour or code a security policy should decide on before install. Notable findings: Collapsed <details> block carrying instructions; Auto-update or dynamic fetch configuration; Auto-update or dynamic fetch configuration. Summary: no network, touches nothing outside the project, writes no memory.
Behavior trace
Honeypot run · 1 s · 2 scenarios · 0 deviationsStatic findings
<summary><b><img src="https://cdn.simpleicons.org/google" width="16" height="16" /> Google Antigravity</b></summary> Export the key before starting `agy` — thatFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.raise RemoteConfigError(f"{name} must be one URL without spaces or control characters")Fix: Pin versions and sources; vendor anything fetched at install time.raise RemoteConfigError(f"{name} has an invalid port") from NoneFix: Pin versions and sources; vendor anything fetched at install time.<summary><b><img src="https://raw.githubusercontent.com/devicons/devicon/master/icons/vscode/vscode-original.svg" width="16" height="16" /> VS Code</b></summaryFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct."access (for example chmod 400 owned by that user)"Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.<summary><b><img src="https://cdn.simpleicons.org/claude" width="16" height="16" /> Claude Desktop</b></summary> Open **Settings > Developer > Edit Config** andFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.raise RemoteConfigError(f"UNIFI_AUDIT_REMOTE_PLAINTEXT=true requires an http:// {name}")Fix: Pin versions and sources; vendor anything fetched at install time.proc = subprocess.Popen( # noqa: S603 - operator-configured argv, never a shellFix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose._BOM = ""Fix: Stop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.from .audit_remote import RemoteConfig, RemoteConfigErrorFix: Pin versions and sources; vendor anything fetched at install time.<summary><b><img src="https://avatars.githubusercontent.com/u/14957082?s=32" width="16" height="16" /> Codex</b></summary> From a terminal (the first line prompFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.<summary><b><img src="https://cdn.simpleicons.org/googlegemini" width="16" height="16" /> Gemini CLI</b></summary> Install this repository as a Gemini CLI extenFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct.raise RemoteConfigError(f"{name} must not carry credentials (user information)")Fix: Pin versions and sources; vendor anything fetched at install time.<summary><b><img src="https://avatars.githubusercontent.com/u/14957082?s=32" width="16" height="16" /> ChatGPT</b></summary> ChatGPT cannot launch a local MCP sFix: Remove the text that addresses the agent rather than the reader; describe behavior, do not instruct._SEPARATORS = "\x00\x01\x02\x03\x04\x05\x06\x07\x0e\x0f\x10\x11"Fix: Put hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.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
Not checked against OSV yet; the weekly check has not reached this version.
What would make this Verified
11 points to 80 · weights: behavior 45%, static 25%, prompt 15%, publisher 15%- Unclaimed publisher≈ +8 on the totalClaim swkstudios: a verified publisher gets +50 on publisher trust. It is free and usually instant with GitHub sign-in.
- Hidden instructions≈ +6 on the totalRemove HTML comments, zero-width characters and white-on-white text that carry instructions. Everything the agent is meant to read should be visible to the person reading the file.
- Auto-update or dynamic fetch configuration≈ +4 on the totalPin versions and sources; vendor anything fetched at install time.
- Hidden zero-width characters detected≈ +4 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.
- Hex-encoded string detected≈ +4 on the totalPut hosts, commands and strings in plain source. Decoding at runtime looks identical to malware and costs points on its own.
- File permission modification command≈ +2 on the totalStop enumerating or reading credential locations; take secrets from a declared environment variable the user sets on purpose.