Static hints overstate the network and child-process uses: inspected code ties them to explicit funding, browser opening, and update commands. The remaining risk is the user-commanded installation of first-party agent skills into a shared agent directory.
Static reason
High-risk behavior combination matched malicious policy.; previous stored version diff introduced dangerous source
Trigger
User runs `tenjin install`.
Impact
Installed skills may influence compatible AI-agent workflows, including marketplace search and publishing.
Mechanism
Copies packaged skills into ~/.agents/skills or ~/.claude/skills.
Evidence
package.jsondist/index.jsdist/install-DXX7PVUP.jsdist/chunk-QWBUKIP2.jsdist/chunk-3ZW7HPY5.jsdist/update-24Y2CQTT.js~/.agents/skills~/.claude/skills
Network endpoints2
tenjin.blogpay.coinbase.com