integram-mcp@0.7.108
MCP server for Integram — connect AI assistants to your workspace via Model Context Protocol
AI Security Review
scanned 4d ago · by lpm-firewall-aiThe source reveals an automatic update and dependency-install capability in Git checkouts, which warrants a warning. The manifest has no install hook, and the inspected code does not establish credential theft, destructive behavior, or another concrete attack.
Decision evidence
public snapshot- package.json declares a CLI entry point and a prepublish-only version check, with no install lifecycle hook.
- When the CLI starts, it checks for updates immediately and repeats the check hourly.
- The updater runs only when the package directory is a Git checkout, pulls from its configured remote, and installs dependencies if the commit changes.
- After an update, it exits so the MCP client can start the updated program.
- The update-check locations flagged as account behavior access a temporary cache and perform the normal login startup; they do not show account changes.
- The automatic updater uses fast-forward-only Git pulls and does not run the npm install step unless the checkout changes.
- The package’s intended service URL is configurable and defaults to localhost; the inspected code did not show credentials sent to an unrelated recipient.
Source & flagged code
3 flagged · loading sourceThis package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
index.jsView on unpkgPackage source invokes a package manager install command at runtime.
index.jsView on unpkg · L3136Findings
Affected versions and remediation
This report applies to integram-mcp@0.7.108.
See all versions of integram-mcp for other recorded verdicts.
- Review the evidence and your use of integram-mcp@0.7.108 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.