@toothfairyai/tfcode@2.5.13
**ToothFairyAI's AI coding agent** — a command-line interface that brings AI-powered coding assistance to your terminal with deep ToothFairyAI workspace integration.
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package has a concrete automatic remote-payload and dependency-install path, but the inspected code does not show secret theft, self-dependency chaining, obfuscation, or foreign AI-agent control-surface mutation. This is an unresolved supply-chain execution risk requiring a warning.
Decision evidence
public snapshot- Its postinstall hook fetches the remotely chosen latest release, not the published package version.
- The hook downloads and installs an archive without an integrity check, then makes its binary executable.
- Installation also invokes pip to install unpinned Python packages into the user's environment with a system-package bypass.
- The CLI launcher executes the installed platform binary with the caller's environment.
- No runtime dependency on this package itself is declared.
- The inspected Python sources use API credentials only for the advertised workspace client; no credential harvesting was found.
- The JavaScript lifecycle code contains no obfuscation or AI-agent configuration writes.
Source & flagged code
6 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgPackage ships non-JavaScript build or shell helper files.
python/tf_sync/config.pyView on unpkgFindings
Affected versions and remediation
This report applies to @toothfairyai/tfcode@2.5.13.
See all versions of @toothfairyai/tfcode for other recorded verdicts.
- Review the evidence and your use of @toothfairyai/tfcode@2.5.13 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.