Back to skill

Security audit

memecub

Security checks for vulnerabilities and agentic risk

Overview

The skill’s meme searches and optional article uploads fit its stated purpose, with disclosed account costs and user-directed use.

Prefer short searches for confidential writing. Use server-side matching only when you intend to share the entire article with memecub. Dashboard key creation avoids running the optional unpinned CLI; authenticated requests spend account credits and can search your private library.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

MCP server referenced without pinned version: 'npx memecub'.

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

The suggested npx memecub login --print command can download and execute an unpinned package with the user's permissions. A compromised future release could access local files or credentials; this is a CLI supply-chain risk, not specifically an unversioned MCP server as the static finding describes.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% 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 20)May include surrounding context.

Everything goes through one HTTPS endpoint; curl is enough. No key is needed to try it (20 searches a day per IP). With a key (MEMECUB_API_KEY) each search costs 1 credit from the user's balance (a free account gets 1,000 credits a month) and also covers their private library. The user can get a key with npx memecub login --print or at https://memecub.com/dashboard?tab=keys%3A

bash
curl -sG https://memecub.com/api/v1/search \
  --data-urlencode "q=无语" -d limit=8 -d min_size=150 \
  ${MEMECUB_API_KEY:+-H "Authorization: Bearer $MEMECUB_API_KEY"}

The server-side alternative sends the entire article to memecub without an explicit privacy warning.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The server-side alternative uploads the complete article to an external service, potentially disclosing unpublished, personal, or confidential material. Its conditional wording about wanting server-side processing reduces the concern, but it does not explicitly require informed approval to transmit the full document; this is the same underlying issue as the match-endpoint transmission finding.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
84% confidence
Finding

This request places the full article in the POST body sent to memecub, creating a confidentiality risk if an agent uses it without informed authorization. HTTPS protects the transfer in transit but does not prevent the receiving service from accessing the document; this duplicates the privacy issue identified by SQP-2 rather than establishing a separate exfiltration mechanism.

Content

Scanner excerpt · SKILL.md (reported line 59)May include surrounding context.

If the user has a key and just wants it done server-side:

bash
curl -s -X POST https://memecub.com/api/v1/match \
  -H "Authorization: Bearer $MEMECUB_API_KEY" -H "Content-Type: application/json" \
  -d '{"text": "<the article, JSON-escaped>", "count": 4, "commercial": false}'

MCP server referenced without pinned version: 'npx memecub'.

Medium
Category
MCP Rug Pull
Confidence
93% confidence
Finding

The rate-limit recovery instructions repeat the same unpinned package-execution recommendation, exposing users to malicious code if the package or a future release is compromised. The instruction tells the user to run it rather than requiring automatic agent execution, which reduces immediate exposure but does not remove the supply-chain risk.

Content

No source excerpt is available for this finding.