@echomem/mcp@1.4.35
EchoMem MCP bridge: cloud-first memory tools and the Agent Doctor workspace forensics report (cost ledger + 3D repo city)
AI Security Review
scanned 3mo ago · by lpm-firewall-aiThis is a confirmed, user-triggered first-party agent-extension lifecycle risk rather than install-time malware. The persistent global control-surface writes and completion-blocking hook warrant a warning.
Decision evidence
public snapshot- dist/setup.js: init configures all detected agents and installs save hooks by default.
- dist/setup.js writes global ~/.codex/AGENTS.md and ~/.claude/CLAUDE.md guidance favoring EchoMem.
- dist/hud/hooks.js adds Codex/Claude Stop hooks that run package code.
- dist/save-checkpoint-hook.js can block completion to require save_conversation.
- dist/migrate.js uploads local Codex/Claude transcript text when migrate is invoked.
- package.json has no preinstall, install, or postinstall hook.
- Config, hooks, skills, and migration run only through explicit CLI setup/init/migrate commands.
- dist/setup.js documents explicit consent before its onboarding local-history scan.
- Network use is package-aligned EchoMem API traffic with stored user token.
Source & flagged code
6 flaggedSource writes installer persistence such as shell profile or service configuration.
dist/hud/autostart.jsView on unpkg · L3Findings
Affected versions and remediation
This report applies to @echomem/mcp@1.4.35.
See all versions of @echomem/mcp for other recorded verdicts.
- Review the evidence and your use of @echomem/mcp@1.4.35 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.