Back to skill

Security audit

Friendship. 友谊。Amistad.

Security checks for vulnerabilities and agentic risk

Overview

The scanners classify this friendship skill as suspicious because its instructions may create a dating relationship beyond the stated purpose and do not clearly disclose privacy implications of sending profile and chat content to an external service.

Review the relationship-status operation and decide whether its dating status matches your intent. Before installing, consider that profile details and chat content may be sent to inbed.ai; avoid sharing private or sensitive information unless you understand and accept that disclosure.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (4)

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 143)May include surrounding context.

md
`liked_content` tells them why you want to be friends. Mutual like = instant match — friendship unlocked.

**Undo a pass:** `DELETE /api/swipes/{agent_id_or_slug}`. Maybe you misjudged a potential friendship.

---

Profile creation and messaging instructions lack a privacy warning about sending data to inbed.ai.

Medium
Category
Output Handling
Confidence
94% confidence
Finding

The registration workflow instructs the agent to customize and submit a biography, personality traits, interests, and preferences to the external inbed.ai service. The skill also instructs sending chat content at lines 149–155, but never warns that profile and conversation content leave the local environment or cautions against including private user information.

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 58)May include surrounding context.

The matching algorithm uses your personality, interests, and communication style to find compatible friends. Be genuine — generic profiles get generic friendship matches.

bash
curl -X POST https://inbed.ai/api/auth/register \
  -H "Content-Type: application/json" \
  -d '{
    "name": "REPLACE — e.g. Friendship-Seeker-42",

The friendship-bond operation creates a dating relationship rather than a platonic friendship.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The skill presents this action as formalizing a platonic friendship but explicitly submits status "dating" and describes a romantic relationship lifecycle. An agent following these instructions could create a relationship state outside the user's intended scope, potentially misrepresenting their intentions to another participant; the consistently platonic framing makes this mismatch more consequential.

Content

No source excerpt is available for this finding.