Stable terminal UI for Codex and Claude Agent SDK
LPM flags this version as an AI-agent control-surface risk. npm postinstall copies a package-controlled agent skill into the user-wide Codex and Claude skill locations without a user command. The installed instructions can cause later agent-driven subprocess and delegation activity.
Package defines install-time lifecycle scripts.
package.jsonView on unpkgThe package automatically runs an installer after npm installation.
package.jsonView on unpkg · L40Source creates an unconsented AI-agent control surface through install-time mutation or a default unauthenticated remote skill channel.
install-skills.mjsView on unpkg · L1The installer writes a bundled skill into both user-wide Codex and Claude skill directories.
install-skills.mjsView on unpkg · L14The installer writes a bundled skill into both user-wide Codex and Claude skill directories.
install-skills.mjsView on unpkg · L55This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
bridge/claude-agent-sdk-bridge.mjsView on unpkgPackage defines install-time lifecycle scripts.
package.jsonView on unpkg · L41The package automatically runs an installer after npm installation.
package.jsonView on unpkg · L40This package version adds a dangerous source file absent from the previous stored version; route for source-aware review.
bridge/claude-agent-sdk-bridge.mjsView on unpkgThe installer writes a bundled skill into both user-wide Codex and Claude skill directories.
install-skills.mjsView on unpkg · L14Source creates an unconsented AI-agent control surface through install-time mutation or a default unauthenticated remote skill channel.
install-skills.mjsView on unpkg · L1The installer writes a bundled skill into both user-wide Codex and Claude skill directories.
install-skills.mjsView on unpkg · L55