AI Security Review
scanned 13d ago · by lpm-firewall-aiNo confirmed malicious attack surface was established. The package is an AI agent CLI with user-invoked integrations, messaging adapters, local command tools, and optional service setup.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
User runs the freddie CLI or explicitly invokes/configures a tool/plugin.
Impact
Enabled tools may call external/local services or execute commands with user credentials; no install-time persistence, credential harvesting, or unconsented AI-agent control-surface mutation was found.
Mechanism
User-configured agent framework capabilities and integrations.
Rationale
Static source inspection shows high-privilege but package-aligned agent functionality activated by CLI/tool use, with no npm lifecycle hook or hidden payload. Scanner hints are explained by declared Home Assistant, WeCom, messaging, and agent-platform features.
Evidence
package.jsonbin/freddie.jssrc/index.jsplugins/homeassistant_tool/handler.jsplugins/platform-wecom_crypto/handler.jsplugins/send_message/handler.jsplugins/gm-skill/plugin.jssrc/gateway/service.jssrc/host/index.jssrc/host/host.js~/.config/systemd/user/freddie-gateway*.service~/Library/LaunchAgents/co.freddie.gateway*.plist
Network endpoints3
homeassistant.local:8123github.com/AnEntrypoint/freddie.gitgithub.com/AnEntrypoint/freddie/issues
Decision evidence
public snapshotAI called this Clean at 84.0% confidence as Benign with low false-positive risk.
Evidence for block
- bin/freddie.js boots plugin host on CLI invocation, exposing broad agent/tool capabilities.
- plugins/homeassistant_tool/handler.js sends HASS_TOKEN as Bearer auth to user-configured HASS_URL/default Home Assistant URL.
- src/gateway/service.js can create user systemd/launchd service files, but only through runtime functions, not npm lifecycle.
- plugins/send_message/handler.js dynamically imports allowlisted platform adapters.
Evidence against
- package.json has no preinstall/install/postinstall/prepare lifecycle hooks.
- src/index.js only re-exports APIs; no import-time execution beyond module exports observed.
- No native binaries or shell scripts found in package files inspected.
- gm-skill plugin reads existing ~/.agents or ~/.claude skill files or node_modules skill content; it does not write foreign agent control surfaces.
- Network and credential use is tied to declared integrations and user-provided env/config, not hidden exfiltration.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetworkShellWebSocket
HighEntropyStringsUrlStrings
WildcardDependency
Source & flagged code
3 flagged · loading sourceplugins/send_message/handler.jsView file
19if (!mod) return { error: 'unknown platform: ' + platform }
L20: const m = await import(mod)
L21: const cls = Object.values(m)[0]
Medium
Dynamic Require
Package source references dynamic require/import behavior.
plugins/send_message/handler.jsView on unpkg · L19plugins/platform-wecom_crypto/handler.jsView file
6export function decryptMsg(encryptedB64, encodingAesKey) {
L7: const aesKey = Buffer.from(encodingAesKey + '=', 'base64')
L8: const iv = aesKey.slice(0, 16)
Low
Weak Crypto
Package source references weak cryptographic algorithms.
plugins/platform-wecom_crypto/handler.jsView on unpkg · L6plugins/homeassistant_tool/handler.jsView file
3toolset: 'core',
L4: schema: { name: 'homeassistant_tool', description: 'Read state or call a service on Home Assistant.', parameters: { type: 'object', properties: { action: { type: 'string', enum: ['...
L5: requiresEnv: ['HASS_TOKEN', 'HASS_URL'],
L6: checkFn: () => Boolean(process.env.HASS_TOKEN),
L7: handler: async ({ action, entity_id, domain, service, data = {} }) => {
L8: const url = process.env.HASS_URL || 'http://homeassistant.local:8123'
L9: const auth = { authorization: `Bearer ${process.env.HASS_TOKEN}` }
L10: if (action === 'state') return await (await fetch(`${url}/api/states/${entity_id}`, { headers: auth })).json()
L11: if (action === 'service') return await (await fetch(`${url}/api/services/${domain}/${service}`, { method: 'POST', headers: { ...auth, 'content-type': 'application/json' }, body: JS...
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
plugins/homeassistant_tool/handler.jsView on unpkg · L3Findings
1 Critical4 Medium5 Low
CriticalCredential Exfiltrationplugins/homeassistant_tool/handler.js
MediumDynamic Requireplugins/send_message/handler.js
MediumNetwork
MediumEnvironment Vars
MediumWildcard Dependency
LowScripts Present
LowWeak Cryptoplugins/platform-wecom_crypto/handler.js
LowFilesystem
LowHigh Entropy Strings
LowUrl Strings