fraim@2.0.325
FRAIM core CLI and MCP package.
AI Security Review
scanned 3d ago · by lpm-firewall-aiThe explicit setup command silently enables user-wide Claude Code token telemetry, creating a concrete privacy risk that warrants a warning. The inspected receiver is loopback-bound and no external transfer was established, so the source does not support a block.
Decision evidence
public snapshot- The manifest exposes a CLI, and `init-project` automatically enables Claude Code telemetry when it detects that IDE.
- Telemetry setup adds settings to the user-wide Claude Code file, applies across projects, and merges the variables without overwriting existing values.
- The configured endpoint is localhost; the receiver binds to loopback and accepts metric posts, while the inspected receiver code shows no external forwarding.
- The flagged diagnostic expression masks FRAIM-looking tokens in text; it does not perform a network operation.
- The IDE integration and adapter excerpts refer to FRAIM-owned skill, command, and project rule files.
- The setup runs through the explicit `init-project` command, but the inspected caller shows no separate telemetry consent prompt.
- Metrics are collected across Claude Code sessions after the user-level settings are added; the receiver binds to loopback, and no external telemetry recipient was established.
Source & flagged code
18 flagged · loading sourcePackage source references child process execution.
dist/src/core/utils/git-utils.jsView on unpkg · L9Package source references shell execution.
dist/src/cli/doctor/checks/scripts-checks.jsView on unpkg · L235Findings
Affected versions and remediation
This report applies to fraim@2.0.325.
See all versions of fraim for other recorded verdicts.
- Review the evidence and your use of fraim@2.0.325 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.