The package performs first-party agent skill setup automatically during installation, creating a real agent-configuration surface despite skip and opt-out controls. The inspected lifecycle source supports a warning, but does not establish a concrete attack.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Package installation outside detected CI, unless automatic setup is disabled or skipped.
Impact
This changes guidance available to a future Claude agent session; the inspected source does not establish credential theft or another active attack.
Mechanism
The postinstall hook reads bundled Ax guidance and writes the skill and its references under .claude/skills.
Evidence
package.jsonscripts/postinstall.mjsscripts/skill-config.mjsscripts/skill-install.mjsskills/ax-llm.md.claude/skills/<skill>/SKILL.md.claude/skills/<skill>/references/