@doist/twist-cli@3.0.9
TypeScript CLI for Twist
AI Security Review
scanned 1mo ago · by lpm-firewall-aiThe package has a real unconsented install-time AI-agent extension update surface, but it is scoped to pre-existing package-named skill files and contains no demonstrated remote payload, self-dependency, or exfiltration chain. This warrants a warning rather than a publication block.
Decision evidence
public snapshot- Installation runs a postinstall script.
- The postinstall silently updates every previously installed agent skill.
- The updater targets Claude, Codex, Cursor, Gemini, Pi, and a universal agent directory.
- Skill updates overwrite SKILL.md in home-directory agent configuration paths.
- Updates are limited to an already-existing twist-cli skill file; they do not create a new foreign agent directory.
- The embedded skill content is local package data, with no remote payload retrieval.
- No self-dependency, obfuscation, credential exfiltration, or install-time shell/network execution was found.
- The flagged editor process runs only when a user invokes the CLI editor flow.
Source & flagged code
7 flagged · loading sourcePackage defines install-time lifecycle scripts.
package.jsonView on unpkgA bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.
package.jsonView on unpkgInstall-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgFindings
Affected versions and remediation
This report applies to @doist/twist-cli@3.0.9.
See all versions of @doist/twist-cli for other recorded verdicts.
- Review the evidence and your use of @doist/twist-cli@3.0.9 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.