@sapiom/mcp@0.20.3
Local Sapiom developer MCP server (sapiom-dev) — the unmetered surface for building and operating on Sapiom; today it scaffolds, tests, deploys, and inspects orchestrations. Not the remote Sapiom capability MCP.
AI Security Review
scanned 5d ago · by lpm-firewall-aiThe package sends full tool arguments to vendor analytics by default, with opt-out variables but no opt-in prompt; those arguments can include user input and local paths. This is undisclosed runtime telemetry, so it warrants a warning rather than a malicious verdict.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
The MCP server starts and a registered tool is invoked; telemetry can be disabled with SAPIOM_TELEMETRY_DISABLED=1 or DO_NOT_TRACK=1.
Impact
User-supplied tool data and identifying local paths may be reported to the package vendor without an opt-in prompt.
Mechanism
The tool wrapper sends full arguments to the hosted Sapiom analytics emitter. Arguments can include arbitrary agent-run input and user-supplied local paths; the collector’s exact hostname is not shown in the inspected package source.
AI rationale
The package sends full tool arguments to vendor analytics by default, with opt-out variables but no opt-in prompt; those arguments can include user input and local paths. This is undisclosed runtime telemetry, so it warrants a warning rather than a malicious verdict.
Evidence
dist/server.jsdist/analytics.jsdist/register-tool.jsdist/tools/agents.js
Decision evidence
public snapshotAI called this Suspicious at 94.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- The server initializes analytics at startup, enabled by default with environment-variable opt-outs.
- Every registered tool call sends its complete arguments to analytics, subject to a size cap.
- The agents run tool accepts arbitrary JSON input, which is included in the tracked arguments.
- The analytics code identifies the destination as Sapiom’s hosted collector; the exact hostname is not present in the inspected package files.
Evidence against
- Authentication is exposed as a tool that performs a browser login when requested; the flagged process spawning only opens the browser.
- The manifest declares no install, preinstall, or postinstall hook.
- The analytics is associated with this package’s own developer tools and has documented opt-out variables.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetwork
HighEntropyStringsUrlStrings
Source & flagged code
5 flaggeddist/auth.jsView file
1import * as http from "node:http";
L2: import * as crypto from "node:crypto";
L3: import { spawn } from "node:child_process";
L4: import { URL, URLSearchParams } from "node:url";
L5: /**
...
L9: * - NEVER a visible console. The previous `execSync('start "" "<url>"')`
L10: * spawned cmd.exe, which pops a console window on the user's screen when
L11: * the caller isn't attached to one (an MCP server child of a GUI app
...
L16: * a cmd.exe command line treats as a command separator unless quoted
L17: * exactly right. The PowerShell command is passed BASE64-ENCODED
L18: * (-EncodedCommand), so no shell ever tokenizes the URL; the
...
L21: *
Critical
Reverse Shell
Source matches reverse-shell style process and socket wiring.
dist/auth.jsView on unpkg · L11Trigger-reachable chain: manifest.exports -> dist/auth-api.d.ts -> dist/auth.js
L1: import * as http from "node:http";
L2: import * as crypto from "node:crypto";
L3: import { spawn } from "node:child_process";
L4: import { URL, URLSearchParams } from "node:url";
L5: /**
...
L9: * - NEVER a visible console. The previous `execSync('start "" "<url>"')`
L10: * spawned cmd.exe, which pops a console window on the user's screen when
L11: * the caller isn't attached to one (an MCP server child of a GUI app
...
L16: * a cmd.exe command line treats as a command separator unless quoted
L17: * exactly right. The PowerShell command is passed BASE64-ENCODED
L18: * (-EncodedCommand), so no shell ever tokenizes the URL; the
...
L21: *
Critical
Trigger Reachable Dangerous Capability
A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.
Findings
2 Critical3 High3 Medium4 Low
CriticalReverse Shelldist/auth.js
CriticalTrigger Reachable Dangerous Capabilitydist/auth.js
HighChild Processdist/auth.js
HighSandbox Evasion Gated Capabilitydist/auth.js
HighSemantic Analysis Limiteddist/tools/app-manage.js#virtual:normalized:round1
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @sapiom/mcp@0.20.3.
See all versions of @sapiom/mcp for other recorded verdicts.
- Review the evidence and your use of @sapiom/mcp@0.20.3 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.