@proagentstore/cli@0.4.199
CLI for creating, publishing, and running ProAgentStore agents
AI Security Review
scanned 2d ago · by lpm-firewall-aiThe CLI automatically uploads local runner stderr output in crash reports to its own service, and the source shows no consent step for that collection. The report is bounded and applies common secret-pattern redaction, so this is suspicious telemetry rather than clear credential theft or malware.
Decision evidence
public snapshot- The runner keeps up to 20 lines of its child process's stderr output in a tail buffer.
- When the local runtime exits unexpectedly, the CLI records that stderr tail in a crash report and attempts delivery using the PAGS session token.
- Crash reports are POSTed to the PAGS API crash-report endpoint; common credential patterns are redacted, but the report includes output text.
- The manifest has no install, preinstall, or postinstall lifecycle scripts.
- The flagged model-list feature sends a provider key to that provider's own model API for the selected engine.
- The package describes itself as a CLI for creating, publishing, and running ProAgentStore agents, consistent with its runner and service-management features.
Source & flagged code
10 flaggedPackage source references child process execution.
dist/browser-runner/site-builder/runtime.jsView on unpkg · L32Package source references shell execution.
dist/browser-runner/coding/windows-runner-live.jsView on unpkg · L59Findings
Affected versions and remediation
This report applies to @proagentstore/cli@0.4.199.
See all versions of @proagentstore/cli for other recorded verdicts.
- Review the evidence and your use of @proagentstore/cli@0.4.199 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.