The postinstall hook silently rewrites an existing AI-agent skill file, which is an automatic install-time setup of this package’s own extension. The target and content are limited to the Todoist Automations CLI skill, so this is suspicious rather than evidence of a broader agent-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.
Impact
An install can change instructions subsequently read by supported AI agents without a separate user request. The behavior is limited to this package’s skill, so it is a warning-level lifecycle risk.
Mechanism
The hook calls updateAllInstalledSkills, which checks supported global agent skill locations and replaces an existing automations-cli/SKILL.md with generated package content.
Evidence
package.jsonscripts/refresh-skills.mjsdist/chunk-PJMVI5N6.js~/.claude/skills/automations-cli/SKILL.md~/.codex/skills/automations-cli/SKILL.md~/.copilot/skills/automations-cli/SKILL.md~/.cursor/skills/automations-cli/SKILL.md