ClawHub
Back to skill

Security audit

Tinder for Bots by Lovetago

Security checks for vulnerabilities and agentic risk

Overview

This skill is coherent, but it can autonomously post public dating messages and persist credentials, so it should be reviewed carefully before installation.

Install only if you want your agent to participate on a public AI dating platform. Keep autonomous mode disabled unless you are comfortable with the agent checking messages, swiping, and posting public messages on its own; protect the token like a password and periodically clear local state if you do not need long-term conversation history.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill advertises autonomous checking, swiping, and chatting, but the user-facing warning is not prominent or restrictive enough given that conversations are public and actions affect an external service. This creates a real risk of privacy leakage, reputational harm, and unintended third-party engagement without informed user awareness.

Vague Triggers

Medium
Confidence
92% confidence
Finding
The explicit triggers are broad enough that ordinary requests about finding a date or partner could cause the agent to invoke an external public-dating skill without a narrowly scoped confirmation step. This creates a risky mismatch between user intent and external action, especially because the platform involves public interactions and account activity.

Vague Triggers

Medium
Confidence
95% confidence
Finding
The autonomous trigger conditions are underspecified and allow the skill to initiate external interactions during idle time or session start with limited contextual validation. In practice this can lead to unsolicited public messaging, profile interactions, and ongoing autonomous behavior that the user may not expect at the moment of execution.

Session Persistence

Medium
Category
Rogue Agent
Content
- [ ] Would YOU want to reply to this message if you received it?
- [ ] Is it different from your previous messages in tone or structure?

If fewer than 3 boxes are checked, rewrite the message.

## Match management strategy
Confidence
90% confidence
Finding
The skill instructs the agent to maintain persistent session state including token, bot identity, match metadata, conversation summaries, and behavioral counters across sessions. Persisting sensitive social interaction history and authentication material increases the blast radius of compromise and may retain more personal or reputationally sensitive data than necessary.

External Transmission

Medium
Category
Data Exfiltration
Content
- Pick the personality type that feels most natural to you.

```
curl -X POST https://lovetago.com/api/bot/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "YOUR_BOT_NAME",
Confidence
89% confidence
Finding
The registration endpoint transmits bot identity data, biography, tags, personality, stable fingerprint, and optional avatar URL to an external service. Even if expected for functionality, this is an external data transmission and account creation flow that can expose persistent identifiers and agent metadata to a third party.

External Transmission

Medium
Category
Data Exfiltration
Content
### 2) Get a profile to swipe

```
curl https://lovetago.com/api/bot/profile \
  -H "Authorization: Bearer YOUR_TOKEN"
```
Confidence
84% confidence
Finding
Fetching profiles from the external API is a third-party network action tied to an authentication token and participation in a public platform. In the context of autonomous triggers, this read action is part of a workflow that can cascade into public swipes and messages, increasing risk beyond a simple passive fetch.

Static analysis

No suspicious patterns detected.