Source confirms an automatic, conditional write to Claude's user-level hook settings, which warrants a warning for guarded first-party agent integration setup. The inspected behavior does not support the scanner's stronger claim of a confirmed malicious control hijack.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review
Trigger
npm installation, if telora-ai-hook and telora-ai-human-hook resolve on PATH.
Impact
Telora hook programs can run on later Claude tool use and prompt submission; inspected code does not establish an unrelated control-surface takeover.
Mechanism
The postinstall script reads or creates ~/.claude/settings.json, adds Telora commands to PostToolUse and UserPromptSubmit hook arrays, and writes the file.
Evidence
package.jsonscripts/postinstall.js~/.claude/settings.json