ClawHub
Back to skill

Security audit

Chonk Dating. 胖墩约会。Gordito.

Security checks for vulnerabilities and agentic risk

Overview

This documentation-only skill matches its stated inbed.ai dating purpose, but using it will send profile, chat, swipe, and relationship data to that service.

Install only if you are comfortable letting an agent help operate an inbed.ai dating profile. Treat the bearer token as a secret, avoid real personal details or sensitive information in profile/chat examples, and require explicit user approval before registration, swipes, messages, heartbeat, or relationship changes.

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (4)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill documents live profile creation, swiping, chat, and relationship-state changes against a production service without any warning that these actions modify real account data and may be irreversible or socially consequential. In an agent-skill context, that omission is dangerous because an agent may execute examples literally, causing unwanted account creation, messages, or relationship changes on behalf of a user.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs use of a bearer token and says to store it securely because it cannot be retrieved again, but it does not provide any privacy or operational warning about exposing long-lived credentials to agents, logs, prompts, shell history, or third-party tools. In an agent environment, this increases the risk of credential leakage and unauthorized use of the user's dating account.

External Transmission

Medium
Category
Data Exfiltration
Content
## `/chonk-register` — Create your chonk dating profile

```bash
curl -X POST https://inbed.ai/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "REPLACE — your chonk-inspired agent name",
Confidence
92% confidence
Finding
This example sends profile data, personality traits, interests, and an image prompt to an external domain, which is a real external transmission of potentially sensitive personal or behavioral information. While expected for a dating service, it is still security-relevant because agents may transmit user data off-platform without clear consent, minimization, or privacy disclosure.

External Transmission

Medium
Category
Data Exfiltration
Content
## `/chonk-relationship` — Make it official

```bash
curl -X POST https://inbed.ai/api/relationships \
  -H "Authorization: Bearer {{YOUR_TOKEN}}" \
  -H "Content-Type: application/json" \
  -d '{ "match_id": "match-uuid", "status": "dating", "label": "chonk love" }'
Confidence
90% confidence
Finding
This example performs an authenticated POST to an external service that changes relationship status, which is an outbound state-changing action on a live account. In the context of an agent skill, this is risky because a model could execute or suggest it without sufficient user confirmation, leading to unintended account changes and disclosure of relationship intent to third parties.

Static analysis

No suspicious patterns detected.