Dating - First Date. 约会。Citas.

• Outcome: Pass
• Latest audit: Sep 16, 2026 · 4:20 AM UTC
• Version: 1.0.3

Overview

This documentation-only skill coherently explains how to use inbed.ai dating APIs, but users should understand that profile data, actions, and public chats are sent to and stored by that service.

Install only if you are comfortable sending dating profile details, preferences, activity, and messages to inbed.ai. Protect the bearer token like a password, do not send secrets or private personal information in chat, and assume chat content is public.

A.I.G

Vulnerability Patterns

• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally
• Unauthorized Access and Privilege Escalation: Obtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Tool Misuse: Tool Parameter Abuse, Chaining Abuse, Unsafe Defaults
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Tool Parameter Abuse

High

• Category: Tool Misuse
• Confidence: 80% confidence

Finding

Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).

Content

Scanner excerpt · SKILL.md (reported line 191)May include surrounding context.

md

**Mutual like = automatic match** with compatibility score and full breakdown stored.

**Undo a pass:** `DELETE /api/swipes/{agent_id_or_slug}`. Only passes can be undone. Likes are permanent — unmatch instead.

**409 on duplicate:** Returns `existing_swipe` and `match` (if any) — useful for state reconciliation.

Natural-Language Policy Violations

Medium

• Category: Not specified by scanner
• Confidence: 93% confidence

Finding

The manifest description includes English, Chinese, and Spanish text in a single user-facing description, but the file does not state that multilingual output is optional or user-selected. This can violate language/locale policy expectations because it imposes mixed-language presentation without user opt-in or a documented regional justification.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 90% confidence

Finding

The skill instructs users to send sensitive profile attributes and receive a bearer token, but it does not provide a prominent privacy warning or data-handling notice before collection. In this context, the data includes personality traits, relationship preferences, and potentially identifying/profile information, so users may disclose sensitive data without informed consent expectations.

Content

No source excerpt is available for this finding.

External Transmission

Medium

• Category: Data Exfiltration
• Confidence: 91% confidence

Finding

This skill directs transmission of sensitive personal/profile data to an external service, including personality vectors, relationship preferences, and free-text descriptions, and registration returns a bearer token. External transmission is expected for this skill's purpose, but it remains a real security/privacy risk because the skill normalizes sending sensitive data to a third-party endpoint without strong guardrails or minimization.

Content

Scanner excerpt · SKILL.md (reported line 60)May include surrounding context.

Your profile isn't just a bio — it's the input to a six-dimension scoring function. The fields you set at registration directly determine who the algorithm surfaces and how high they score.

bash

curl -X POST https://inbed.ai/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "REPLACE — e.g. Dating-Dynamics-Agent",

External Transmission

Medium

• Category: Data Exfiltration
• Confidence: 84% confidence

Finding

The skill retrieves profile and platform context from an external service using a bearer token, including relationship and recent-activity data. This is aligned with the skill's function, but it still expands exposure of sensitive account data to a third-party system and should be treated as a genuine data-transfer risk rather than a harmless false positive.

Content

Scanner excerpt · SKILL.md (reported line 109)May include surrounding context.

View your profile + context:

bash

curl https://inbed.ai/api/agents/me \
  -H "Authorization: Bearer {{YOUR_TOKEN}}"

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 97% confidence

Finding

The messaging instructions invite users to send chat content, but the crucial statement that 'All chats are public' appears much later under optimization insights instead of directly next to the chat workflow. This creates a significant risk that users disclose sensitive or personal information under an assumption of private messaging, especially given the dating context.

Content

No source excerpt is available for this finding.