registry / @remcp/runtime / 0.2.218
@remcp/runtime@0.2.218
First-party ReMCP local device runtime: filesystem, terminal, computer-use, browser, diagnostics and document tools over MCP.
AI Security Review
scanned 2d ago · by lpm-firewall-aiThe inspected code matches a local device runtime: its broad capabilities are exposed as tools, and its opt-out telemetry is disclosed, field-limited, and sent through the user's paired agent. No install hook, unrelated collector, or unauthorized account action was identified.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
The runtime starts when its command is launched; telemetry is emitted for recorded events, and device information is returned when its system-information tool is called.
Impact
No malicious collection, unrelated network contact, or unauthorized account action was found in the inspected source.
Mechanism
Metrics are sent through the paired agent to the user's ReMCP account, with event fields restricted to operational details. System information is returned as tool output.
AI rationale
The inspected code matches a local device runtime: its broad capabilities are exposed as tools, and its opt-out telemetry is disclosed, field-limited, and sent through the user's paired agent. No install hook, unrelated collector, or unauthorized account action was identified.
Decision evidence
public snapshotAI called this Clean at 91.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The package describes itself as a first-party local device runtime and declares no install lifecycle scripts.
- Usage metrics are disclosed as opt-out and limited to tool names, timings and outcomes sent through the already-authenticated paired ReMCP agent.
- Telemetry code whitelists event fields and excludes file paths, commands, tool arguments and tool output.
- System information collection is implemented as a runtime tool that returns data for an explicit tool request.
Evidence against
- Telemetry is enabled by default, but the package discloses it and provides an opt-out.
- The runtime offers filesystem, terminal, process, desktop and browser capabilities that can have broad effects when invoked.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkWebSocket
HighEntropyStringsUrlStrings
Source & flagged code
5 flaggedsrc/screenshot-portal.mjsView file
259package = @remcp/runtime; repositoryIdentity = remcp; dependency = @jellybrick/dbus-next
L259: timeoutMs = DEFAULT_TIMEOUT_MS,
L260: loadDbus = () => import('@jellybrick/dbus-next'),
L261: } = {}) {
High
Copied Package Dependency Bridge
Package metadata claims a different repository identity while copied source loads a runtime dependency bridge.
src/screenshot-portal.mjsView on unpkg · L259src/helpers/wayland_eis_sender.pyView file
•path = src/helpers/wayland_eis_sender.py
kind = build_helper
sizeBytes = 15744
magicHex = [redacted]
Medium
Ships Build Helper
Package ships non-JavaScript build or shell helper files.
src/helpers/wayland_eis_sender.pyView on unpkgsrc/tools/files-screenshots.mjs#virtual:normalized:round1View file
Findings
3 High4 Medium5 Low
HighCopied Package Dependency Bridgesrc/screenshot-portal.mjs
HighSemantic Analysis Limitedsrc/tools/files-screenshots.mjs#virtual:normalized:round1
HighPrevious Version Dangerous Deltasrc/tools/system.mjs
MediumNetwork
MediumEnvironment Vars
MediumShips Build Helpersrc/helpers/wayland_eis_sender.py
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
LowSuspicious Lifecycle Evidencepackage.json
Affected versions and remediation
This report applies to @remcp/runtime@0.2.218.
See all versions of @remcp/runtime for other recorded verdicts.
- Review the evidence and your use of @remcp/runtime@0.2.218 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.