@echomem/mcp@1.4.9
EchoMem MCP bridge: cloud-first memory tools, local context HUD, and the Agent Doctor workspace forensics report (cost ledger + 3D repo city)
AI Security Review
scanned 3mo ago · by lpm-firewall-aiSource inspection found no malicious install-time behavior, but it provides user-invoked mutations of global AI-agent configuration and guidance. Flag as a warning for this capability rather than block it.
Decision evidence
public snapshot- dist/setup.js: explicit setup/init writes MCP entries and global agent guidance
- dist/setup.js: guidance tells Codex/Claude to call EchoMem memory tools
- dist/hud/hooks.js: explicit install-hooks writes ~/.codex/hooks.json commands
- dist/hud/autostart.js: explicit CLI action creates a macOS LaunchAgent
- package.json has no preinstall/install/postinstall lifecycle hook
- dist/setup.js gates configuration writes behind setup/init commands
- dist/hud/cli.js gates hooks and autostart behind named user commands
- dist/index.js sends MCP memory requests to the documented EchoMem API with its stored token
- dist/keystore.js stores credentials locally with restrictive modes
Source & flagged code
7 flaggedA manifest entrypoint or package-local install chain reaches persistence behavior.
dist/hud/autostart.jsView on unpkg · L3Findings
Affected versions and remediation
This report applies to @echomem/mcp@1.4.9.
See all versions of @echomem/mcp for other recorded verdicts.
- Review the evidence and your use of @echomem/mcp@1.4.9 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.