@uipath/cli@1.202.0
Cross platform CLI for UiPath
AI Security Review
scanned 28d ago · by lpm-firewall-aiAgent skill and Claude plugin changes are a real first-party capability, but they run only from explicit CLI commands and not from npm lifecycle hooks. That matches a user-invoked agent configuration warning rather than an unconsented install hijack.
Decision evidence
public snapshot- The skills install command copies UiPath skills into selected coding-agent directories and, for Claude, runs the Claude CLI to register a local marketplace and install the uipath plugin.
- package.json publishes the uip bin and defines no preinstall, install, or postinstall scripts.
- Skill and plugin refresh runs only when the user invokes uip skills update or uip update.
- Agent directory writes and the Claude plugin install are reached only from explicit skills or update subcommands.
- The Claude plugin id is the vendor plugin uipath on marketplace uipath-marketplace.
- No install hook, eval payload, or unrelated credential send was found on this command path.
Source & flagged code
14 flagged · loading sourceThis package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
dist/applicationinsights-s28k1t0d.jsView on unpkgPackage source references child process execution.
dist/applicationinsights-s28k1t0d.jsView on unpkg · L11246Findings
Affected versions and remediation
This report applies to @uipath/cli@1.202.0.
See all versions of @uipath/cli for other recorded verdicts.
- Review the evidence and your use of @uipath/cli@1.202.0 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.