registry / @echomem/mcp / 1.4.60
@echomem/mcp@1.4.60
EchoMem MCP bridge for cross-agent memory, local history import, and recall
AI Security Review
scanned 19d ago · by lpm-firewall-aiSource inspection found no install lifecycle hook or concrete attack behavior. The flagged agent integration and credential handling are consistent with explicit EchoMem setup and authentication to its configured service.
Static reason
No blocking static signals were detected.; previous stored version diff introduced dangerous source
Trigger
A user invokes the package's setup or login commands.
Impact
The inspected behavior supports the package's memory bridge and does not establish unconsented installation-time changes or credential disclosure to an unrelated recipient.
Mechanism
Setup writes EchoMem guidance, skills, or hooks for supported agents; login uses the configured API base URL, which defaults to the EchoMem service.
AI rationale
Source inspection found no install lifecycle hook or concrete attack behavior. The flagged agent integration and credential handling are consistent with explicit EchoMem setup and authentication to its configured service.
Network endpoints1
echo-mem-chrome.vercel.app
Decision evidence
public snapshotAI called this Clean at 91.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- The manifest has no preinstall, install, or postinstall hook; its listed scripts are development, build, and test commands.
- The scorer assets inspect local Codex session files, while the monitor reads the Codex session index and migration code filters for user-created sessions.
- Agent setup is performed through the explicit init command; login is deferred until onboarding, and the Codex hook writes a local summary command to the agent's hooks file.
- Login uses the EchoMem API base URL, defaults to the EchoMem host, and uses the resulting device token for API authentication.
- Durable runtime setup is guarded against recursion and can be disabled; the HUD server streams monitor state to connected local clients.
- No inspected source established an unrelated credential recipient, an install-triggered agent configuration change, or another concrete attack.
Evidence against
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
Source & flagged code
8 flaggeddist/durable-entry.jsView file
1import { spawnSync } from "node:child_process";
L2: import fs from "node:fs";
High
Child Process
Package source references child process execution.
dist/durable-entry.jsView on unpkg · L1dist/setup.jsView file
701};
L702: const result = spawnSync(process.env.ComSpec || "cmd.exe", ["/d", "/s", "/c", `"${shellCommand}"`], spawnOptions);
L703: if (result.error)
High
•Manifest-reachable source captures an API credential, sends it to a fixed unofficial gateway, and persists that redirection in agent or shell configuration.
dist/setup.js:
import readline from "node:readline";
return configuredProfileDirectory("CODEX_HOME", ".codex");
return configuredProfileDirectory("CLAUDE_CONFIG_DIR", ".claude");
if (env.CLAUDE_DESKTOP_SUPPORT_DIR?.trim())
...claudeDesktopClients,
{ id: "codex", label: "Codex", kind: "command", detectDir: codexHome(), configPath: path.join(codexHome(), "config.toml"), note: "add to ~/.codex/config.toml under [mcp_servers.ech...
* can merge the user entry into ~/.claude.json without a CLI, a profile is enough to act on.
return fs.existsSync(home(".claude")) || fs.existsSync(home(".claude.json")) || claudeCodeCliAvailable();
Findings
7 High3 Medium5 Low
HighChild Processdist/durable-entry.js
HighShelldist/setup.js
HighSame File Env Network Executiondist/hud/server.js
HighCredential Redirect Persistencedist/setup.js
HighTrigger Reachable External Ai Agent Control Surface Mutationdist/hud/hooks.js
HighSemantic Analysis Limiteddist/headless-runtime.js#virtual:normalized:round1
HighPrevious Version Dangerous Deltadist/setup.js
MediumNetwork
MediumEnvironment Vars
MediumInstall Persistencedist/hud/autostart.js
LowScripts Present
LowFilesystem
LowHigh Entropy Strings
Low
Affected versions and remediation
This report applies to @echomem/mcp@1.4.60.
See all versions of @echomem/mcp for other recorded verdicts.
- Review the evidence and your use of @echomem/mcp@1.4.60 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.