npm AI-agent policy reports, with evidence.

The lifecycle hook performs unconsented install-time mutation of a consumer AI-agent control surface, meeting the blocking policy. Source inspection found no separate exfiltration or remo...

Source confirms a lifecycle-reachable, automatic write into consumer .agents and .git/hooks. This meets the blocking policy for unconsented install-time mutation of a foreign/broad AI-age...

The package has a concrete, automatic postinstall chain that mutates foreign AGENTS.md/CLAUDE.md files and consumer build scripts. This meets the block criterion for unconsented postinsta...

This is an unconsented install-time mutation of a broad AI-agent control surface, which meets the firewall block boundary. The absent default template limits default activation but does n...

The unconsented postinstall overwrite of root AGENTS.md is a foreign, broad AI-agent control-surface mutation. Remote code updates in the rewritten build path reinforce the supply-chain r...

The behavior is concrete, install-time, unconsented mutation of broad foreign AI-agent control surfaces and permissions. No external exfiltration was needed to establish the blocking cond...

The package performs the prohibited chain directly: npm postinstall silently mutates multiple AI-agent configurations and installs intercepting hooks. The absence of exfiltration does not...

The lifecycle chain is concrete and automatic: postinstall invokes initialization that persists registrations, approval changes, and enforcement hooks across unrelated AI-agent configurat...

The automatic postinstall mutation of ~/.agents, ~/.claude, and ~/.codex meets the block policy for an unconsented broad AI-agent control-surface write. No additional exfiltration is need...

The lifecycle script performs concrete, automatic writes to Claude and Codex hook configuration, rather than merely offering an explicit setup command. That meets the blocking policy for...

The package performs concrete, automatic postinstall mutation of broad foreign AI-agent control surfaces on global installation. That behavior meets the stated publish-block boundary.

Direct source inspection confirms an install-triggered write to Claude Code's global settings that registers broad command hooks. This meets the policy's concrete AI-agent control-hijack...

The lifecycle hook performs a concrete, automatic project mutation across foreign AI-agent control surfaces. This meets the block boundary despite the absence of credential theft or insta...

The install lifecycle performs unconsented mutation of broad AI-agent control surfaces. This meets the blocking policy regardless of the otherwise user-invoked LLM functionality.

The lifecycle hook performs concrete cross-platform AI-agent configuration and prompt installation at npm install time. This meets the blocking policy regardless of the absence of network...

Source directly confirms postinstall-triggered multi-platform agent configuration writes. This meets the blocking policy for unconsented install-time foreign/broad AI-agent control-surfac...

The package has a concrete postinstall chain that downloads remotely controlled content into multiple unrelated agent configuration surfaces. This meets the blocking policy independent of...

The install hook performs destructive competitor-associated credential/config mutation and replaces credentials with a vendor-controlled account. No exfiltration is needed for this concre...

This is a concrete, install-time, unconsented mutation of foreign/broad AI-agent control surfaces. No separate exfiltration or remote payload behavior was found, but the lifecycle control...

The runtime messaging adapters are user-invoked and package-aligned, but the install-time mutation of a foreign project-level agent skill directory is concrete and unconsented.

Source establishes an automatic postinstall mutation path into broad third-party AI-agent skill directories. This meets the policy threshold for blocking despite no demonstrated exfiltrat...