crescodb@1.2.0
Your whole backend in a few commands: auto-generated REST API, auth with roles, file storage and realtime over SQLite, PostgreSQL or MySQL. Runs locally, deploys to a server you own, and ejects in one command.
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe flagged deploy upload is user-triggered and excludes `.env` files, so it does not establish automatic exfiltration. However, Cloud deployments automatically enable runtime monitoring, and the package README does not disclose that reporting; this warrants a telemetry warning.
Static reason
One or more suspicious static signals were detected.; previous stored version diff introduced dangerous source
Trigger
Production server startup with CRESCO_MONITOR_TOKEN; Cresco Cloud deployments inject this token.
Impact
The platform receives runtime and deployment details and collected application metrics. The package README does not explain this reporting, so Cloud users may not expect it.
Mechanism
A background timer posts monitoring reports to the configured URL's /monitor/ingest endpoint, defaulting to the Cresco platform.
AI rationale
The flagged deploy upload is user-triggered and excludes `.env` files, so it does not establish automatic exfiltration. However, Cloud deployments automatically enable runtime monitoring, and the package README does not disclose that reporting; this warrants a telemetry warning.
Evidence
src/cli/prod.jssrc/server/monitor.js
Network endpoints1
api.crescodb.com/monitor/ingest
Decision evidence
public snapshotAI called this Suspicious at 83.0% confidence as Dangerous Capability with medium false-positive risk.
Evidence for warning
- Production monitoring uses CRESCO_MONITOR_TOKEN, which the source says Cresco Cloud injects automatically.
- With a token configured, a background timer sends reports to the configured monitoring URL, defaulting to the Cresco platform.
- The report includes runtime and deployment metadata along with collected application aggregates; the package README does not describe this runtime reporting.
Evidence against
- Without a monitoring token, the monitor does not start; self-hosters must provide one.
- The deploy upload is reached only when the user runs `cresco deploy --server`; it excludes `.env` files and local databases.
- The monitored data is sent to a monitoring endpoint and the feature is identified in source comments as production monitoring.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShell
HighEntropyStrings
Source & flagged code
10 flaggedsrc/server/manifest.jsView file
41patternName = generic_password
severity = medium
line = 41
Medium
42patternName = generic_password
severity = medium
line = 42
Medium
src/core/agent-exec.jsView file
6L7: import { spawn } from 'child_process';
L8: import path from 'path';
High
Child Process
Package source references child process execution.
src/core/agent-exec.jsView on unpkg · L6Findings
1 Critical5 High5 Medium5 Low
CriticalPrevious Version Dangerous Deltasrc/cli/eject.js
HighChild Processsrc/core/agent-exec.js
HighShellsrc/cli/install.js
HighCommand Output Exfiltrationsrc/cli/deploy.js#virtual:normalized:round1
HighTrigger Reachable Command Output Exfiltrationsrc/cli/deploy.js#virtual:normalized:round1
HighSemantic Analysis Limitedsrc/cli/deploy.js#virtual:normalized:round1
MediumSecret Patternsrc/server/manifest.js
MediumDynamic Requirebin/cresco.js
MediumNetwork
MediumEnvironment Vars
MediumSecret Patternsrc/server/manifest.js
LowScripts Present
Affected versions and remediation
This report applies to crescodb@1.2.0.
See all versions of crescodb for other recorded verdicts.
- Review the evidence and your use of crescodb@1.2.0 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.