The package’s postinstall automatically installs its own skill into global AI-agent directories and replaces an existing copy. This is an agent extension lifecycle risk, but the bundled skill is the package’s own and disables model invocation by default, so the evidence supports a warning rather than a block.
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
Runs automatically during npm postinstall.
Impact
Adds the package’s skill to the user’s agent configuration and may replace local changes to a same-named skill.
Mechanism
The CLI install command copies bundled skills to the user’s Claude and agent skill directories, removing each destination first.
Evidence
package.jsondist/cli.jsskills/initializing-context-repo/SKILL.md~/.claude/skills/initializing-context-repo~/.agents/skills/initializing-context-repo