@desplega.ai/agent-fs@0.15.4
Agent-first filesystem backed by S3
AI Security Review
scanned 25d ago · by lpm-firewall-aiThe scanner’s credential-exfiltration and shell-execution labels are not supported by the inspected excerpts: telemetry uses anonymous product event data, and the child-process import is associated with explicit onboarding. The manifest shows only a prepublish-only build hook, so the source reviewed does not establish malicious behavior.
Decision evidence
public snapshot- The package has a prepublish-only build hook; it does not declare an install-time lifecycle hook.
- The CLI’s telemetry code supports opt-out through ANONYMIZED_TELEMETRY or DO_NOT_TRACK.
- Telemetry is sent as a POST to a fixed endpoint; inspected code constructs anonymous product event data.
- The CLI also contains child-process use in its explicit onboarding command.
- The inspected telemetry payload uses anonymous event metadata and aggregate usage counts, not environment variables, credentials, or command output.
- No concrete malicious behavior was established in the inspected CLI excerpts.
Source & flagged code
8 flaggedA single source file combines environment access, network access, and code or shell execution with blocking evidence.
dist/cli.jsView on unpkg · L292Findings
Affected versions and remediation
This report applies to @desplega.ai/agent-fs@0.15.4.
See all versions of @desplega.ai/agent-fs for other recorded verdicts.
- Review the evidence and your use of @desplega.ai/agent-fs@0.15.4 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.