registry / @mnemonik/claude-code-hooks / 0.9.23

@mnemonik/claude-code-hooks@0.9.23

Claude Code lifecycle contextual hooks for Mnemonik. PreToolUse Edit/Write/NotebookEdit/Bash gates plus SessionStart bootstrap warming, clean Stop, PreCompact context_snapshot creation, bounded SessionEnd cleanup, and UserPromptSubmit context injection. T

AI Security Review

scanned 21h ago · by lpm-firewall-ai

Review flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.

Static reason

No blocking static signals were detected.

Trigger

User runs `mnemonik-claude-code-hooks install`; subsequently configured Claude Code hook events run the dispatcher.

Impact

Can influence Claude Code tool decisions and transmit submitted prompts, tool arguments, paths, session IDs, and context statistics to the configured Mnemonik server.

Mechanism

User-invoked Claude Code configuration and authenticated hook telemetry/context delivery.

Rationale

Source inspection found a first-party, explicit-user-command Claude Code extension installer, not an unconsented npm lifecycle mutation. Per policy it warrants a warning for its persistent agent-control surface, while evidence does not support malicious intent.

Evidence

package.jsondist/install.jsdist/hook.jsdist/lib/http.jsdist/lib/env.jsdist/lib/installerSupport.js~/.claude/settings.json.claude/settings.json~/.mnemonik/hooks/claude-code-hooks/~/.mnemonik/hooks/claude-code-hooks/session-incarnations/~/.mnemonik/scanner.json.mnemonik.json

Network endpoints5

@mnemonik/claude-code-hooks@0.9.23

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code