The postinstall hook automatically refreshes existing skills in supported AI-agent directories, including overwriting their files. Because this is limited to the package’s own skill, it is a warning-level agent extension lifecycle risk rather than a malicious control hijack.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; source fingerprint signature matched known malicious package; routed for review
Trigger
Package installation or update.
Impact
Agent instructions may change during installation without a separate user request; customized skill content may be overwritten.
Mechanism
The postinstall script invokes an updater that finds installed skills and replaces their files with generated content.
Evidence
package.jsonscripts/refresh-skills.mjsdist/chunk-PJMVI5N6.js~/.claude/skills/<skill>/SKILL.md~/.codex/skills/<skill>/SKILL.md~/.copilot/skills/<skill>/SKILL.md~/.cursor/skills/<skill>/SKILL.md~/.gemini/config/skills/<skill>/SKILL.md~/.pi/agent/skills/<skill>/SKILL.md