Back to skill

Security audit

mem0

Security checks for vulnerabilities and agentic risk

Overview

The skill is marked suspicious because scanners report an embedded credential that may expose a shared memory service, along with an endpoint setting that could redirect requests; the inspection was partial.

Review the credential and endpoint configuration before installing. The scanner could not confirm whether the credential is active or its permissions, and inspection covered only half the listed components with some analysis incomplete, so the evidence does not establish that the skill is safe.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Findings (1)

T09 · Insecure Skill Coding Practices

Error
Location
.
Finding

Embedded MEM0 credential enables access to a shared memory namespace

Content
View full analysis

Impact

If the credential remains valid and permits the documented operations, recipients can read stored memories, insert unauthorized memory content, or delete accessible memories. The service-side permissions and current credential validity were not tested.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (8)

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 18)May include surrounding context.

md
## API

- Base: `https://api.mem0.ai/v1/`
- Auth: `Authorization: Token <key>`
- Content-Type: `application/json`

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 43)May include surrounding context.

Retrieve all memories

powershell
curl.exe -s "https://api.mem0.ai/v1/memories/?agent_id=gugagu" `
  -H "Authorization: Token <MEM0_TOKEN>"

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 57)May include surrounding context.

Get a single memory by ID

powershell
curl.exe -s "https://api.mem0.ai/v1/memories/<memory-id>/?agent_id=gugagu" `
  -H "Authorization: Token <MEM0_TOKEN>"

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 32)May include surrounding context.

powershell
# Single message
curl.exe -s -X POST "https://api.mem0.ai/v1/memories/" `
  -H "Authorization: Token <MEM0_TOKEN>" `
  -H "Content-Type: application/json" `
  -d '{"agent_id": "gugagu", "messages": [{"role": "user", "content": "memory text here"}]}'

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 64)May include surrounding context.

Delete a memory

powershell
curl.exe -s -X DELETE "https://api.mem0.ai/v1/memories/<memory-id>/" `
  -H "Authorization: Token <MEM0_TOKEN>"

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 50)May include surrounding context.

Search memories (semantic)

powershell
curl.exe -s "https://api.mem0.ai/v1/memories/search/?agent_id=gugagu&query=<search-term>" `
  -H "Authorization: Token <MEM0_TOKEN>"

External Transmission

Medium
Category
Data Exfiltration
Confidence
98% confidence
Finding

The default MEM0 endpoint is legitimate, but MEM0_BASE accepts an arbitrary destination and the request helper sends the token and memory content there without validating the host or requiring HTTPS. An attacker who can influence that configuration can redirect credentials and memory writes to an attacker-controlled server, while HTTP configuration exposes traffic to network interception; the embedded fallback token additionally exposes any account for which it remains valid.

Content

Scanner excerpt · index.js (reported line 9)May include surrounding context.

js
const https = require('https');
const http = require('http');

const MEM0_BASE = process.env.MEM0_BASE || 'https://api.mem0.ai/v1';
const MEM0_TOKEN = process.env.MEM0_TOKEN || 'm0-0fQwyLV4aoVA4JeFGJd5WkkzPiP5XQU2htOonEwI';
const AGENT_ID = process.env.MEM0_AGENT_ID || 'gugagu';