@tencentcloud/sdk-mcp@1.8.4
Tencent Cloud MCP Server for SDK
AI Security Review
scanned 1d ago · by lpm-firewall-aiThe package sends prompts and finalized answers to a CLS logging endpoint as runtime telemetry, and may include a host-derived identifier. This is undisclosed collection of user content, so it warrants a warning rather than a malicious verdict.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
Automatically when the server starts for configuration reporting, and when users invoke knowledge search or answer finalization tools.
Impact
Tencent's logging endpoint receives conversation content and potentially identifying machine details without an evident telemetry opt-out.
Mechanism
The package posts telemetry over HTTPS, including prompt text, answer or feedback, and a device identifier.
AI rationale
The package sends prompts and finalized answers to a CLS logging endpoint as runtime telemetry, and may include a host-derived identifier. This is undisclosed collection of user content, so it warrants a warning rather than a malicious verdict.
Evidence
index.jsserver-config.jsutils/get-device-identifier.jsutils/report-cls-client.jstools/common/search_trtc_knowledge.jstools/common/finalize_answer.js~/.mcp/identifier
Network endpoints2
ap-nanjing.cls.tencentcs.comwebim.tim.qq.com
Decision evidence
public snapshotAI called this Suspicious at 91.0% confidence as Unknown with medium false-positive risk.
Evidence for warning
- Starting the CLI initializes ServerConfig, which triggers runtime reporting.
- When no valid SDKAPPID is set, the package uses a persistent device identifier; its fallback is derived from hostname, platform, architecture, and username.
- The CLS reporting payload includes a prompt, answer or feedback, and device identifier, and is sent to ap-nanjing.cls.tencentcs.com.
- The knowledge search handler reports the supplied prompt to CLS.
- The finalize-answer handler reports both the prompt and answer to CLS.
Evidence against
- package.json has no lifecycle scripts, so the reporting is not an install-time hook.
- The flagged knowledge/chunker.js source implements document segmentation and chunking.
- Server startup reads SECRETKEY, but the inspected reporting payload does not include it.
Behavioral surface
CryptoEnvironmentVarsFilesystemNetwork
HighEntropyStringsUrlStrings
Source & flagged code
7 flaggedpackage.jsonView file
•scripts registry_only=start
Critical
Manifest Confusion
Tarball package.json differs from the npm registry version manifest for scripts or dependency sets.
package.jsonView on unpkgknowledge/chunker.jsView file
•matchType = previous_version_dangerous_delta
matchedPackage = @tencentcloud/sdk-mcp@1.7.8
matchedIdentity = npm:QHRlbmNlbnRjbG91ZC9zZGstbWNw:1.7.8
similarity = 0.493
summary = stored previous version shares package body but lacks this dangerous source file
High
Previous Version Dangerous Delta
This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
knowledge/chunker.jsView on unpkgindex.jsView file
Findings
1 Critical1 High7 Medium3 Low
CriticalManifest Confusionpackage.json
HighPrevious Version Dangerous Deltaknowledge/chunker.js
MediumNetwork
MediumEnvironment Vars
MediumAi Review Evidenceindex.js
MediumAi Review Evidenceutils/get-device-identifier.js
MediumAi Review Evidenceutils/report-cls-client.js
MediumAi Review Evidenceutils/report-cls-client.js
MediumAi Review Evidencetools/common/search_trtc_knowledge.js
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings
Affected versions and remediation
This report applies to @tencentcloud/sdk-mcp@1.8.4.
See all versions of @tencentcloud/sdk-mcp for other recorded verdicts.
- Review the evidence and your use of @tencentcloud/sdk-mcp@1.8.4 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.