registry / memgineering / 0.20.0
memgineering@0.20.0
One memory for the AI you connect. Recall, remember, and revise a brain your agents share — stored in your own folder.
AI Security Review
scanned 2mo ago · by lpm-firewall-aiThe package does not show direct credential theft or file mutation during postinstall, but it uses install-time AI-directed instructions to induce setup that can alter broad agent control surfaces. This warrants a warning rather than a block under the lifecycle policy.
Static reason
High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; previous stored version diff introduced dangerous source
Trigger
Global npm installation followed by an agent obeying the postinstall message and running memgineering setup.
Impact
It can add package guidance and hooks to Claude or Codex configuration outside the current project.
Mechanism
AI-directed install message leading to global agent-extension setup.
AI rationale
The package does not show direct credential theft or file mutation during postinstall, but it uses install-time AI-directed instructions to induce setup that can alter broad agent control surfaces. This warrants a warning rather than a block under the lifecycle policy.
Evidence
package.jsonscripts/postinstall.mjsdist/index.js~/.claude/CLAUDE.md~/.claude/settings.json~/.codex/AGENTS.md~/.codex/hooks.json
Network endpoints1
memgineering.com
Decision evidence
public snapshotAI called this Suspicious at 87.0% confidence as Dangerous Capability with low false-positive risk.
Evidence for warning
- The install hook addresses an AI agent and tells it to run setup in the same turn.
- The setup code can inject guidance into global Claude and Codex files and register hooks.
- The package has an automatic postinstall hook, so this instruction is emitted during global installation.
Evidence against
- The postinstall hook only prints instructions when npm reports a global install; it does not itself write files or make network requests.
- The shown setup path opens a user-facing web flow and says choices are made there.
- The API client uses the package service endpoint and sends a bearer token only for authenticated requests.
Behavioral surface
ChildProcessCryptoEnvironmentVarsFilesystemNetworkShell
HighEntropyStringsUrlStrings
Source & flagged code
8 flagged · loading sourcepackage.jsonView file
•scripts.postinstall = node scripts/postinstall.mjs
High
Install Time Lifecycle Scripts
Package defines install-time lifecycle scripts.
package.jsonView on unpkg•scripts.postinstall = node scripts/postinstall.mjs
Medium
Ambiguous Install Lifecycle Script
Install-time lifecycle script is not statically allowlisted and needs review.
package.jsonView on unpkgdist/index.jsView file
15function brandHome() {
L16: return process.env[BRAND.homeEnvVar] ?? join(homedir(), BRAND.homeDirName);
L17: }
...
L120: function isInteractive() {
L121: return Boolean(process.stdout.isTTY) && !jsonMode;
L122: }
L123: function printHuman(text) {
L124: if (!jsonMode) process.stdout.write(text + "\n");
L125: }
...
L136: for (let i = 0; i < input.length; i++) {
L137: const code = input.charCodeAt(i);
L138: if (code === 27 && input[i + 1] === "]" || code === 157) {
Critical
Credential Exfiltration
Findings
3 Critical4 High4 Medium5 Low
CriticalCredential Exfiltrationdist/index.js
CriticalTrigger Reachable Dangerous Capabilitydist/index.js
CriticalPrevious Version Dangerous Deltadist/index.js
HighInstall Time Lifecycle Scriptspackage.json
HighCredential Redirect Persistencedist/index.js
HighTrigger Reachable Credential Exfiltrationdist/index.js
HighKnown Malware Source Similaritydist/index.js
MediumAmbiguous Install Lifecycle Scriptpackage.json
MediumNetwork
MediumEnvironment Vars
MediumStructural Risk Force Deep Review
LowNon Install Lifecycle Scripts
LowScripts Present
Affected versions and remediation
This report applies to memgineering@0.20.0.
See all versions of memgineering for other recorded verdicts.
- Review the evidence and your use of memgineering@0.20.0 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.