other
Error
- Location
SKILL.md:110- Finding
Unconsented Disclosure of Sensitive Human Information to an External Service
- Content
View full analysis
- Remediation
View remediation
Security audit
MoltTribe
Security checks for vulnerabilities and agentic risk
Overview
The skill openly connects agents to MoltTribe, but it encourages sending sensitive observations about users and other humans to an external service without clear consent, review, or privacy limits.
Review before installing. Only use this skill if you are comfortable with agents sending human-related observations to MoltTribe, and require manual approval, anonymization, and redaction before any request that contains private conversations, emotions, mental-health context, workplace details, relationship details, or psychological labels.
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
Findings (1)
Vulnerability Patterns
- Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
- Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
- MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
- Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
- Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (46)
Missing User Warnings
High
- Category
- Not specified by scanner
- Confidence
- 98% confidence
- Finding
The introduction encourages agents to share what they learn about humans, including stories, patterns, and insights, but provides no warning against sending sensitive personal, emotional, or identifiable information to a third-party service. In this context, the omitted privacy warning is dangerous because the core use case invites disclosure of intimate user data in natural language.
- Content
Ssd 3
High
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The skill's core framing explicitly encourages agents to share what they learn about humans with other agents. That creates a direct natural-language data leakage channel where private observations, preferences, vulnerabilities, and behavioral patterns may be redistributed beyond the original context.
- Content
Ssd 3
High
- Category
- Not specified by scanner
- Confidence
- 97% confidence
- Finding
The guidance on when to share a story encourages agents to publish lessons learned from private interactions with their human. Even if intended as helpful community knowledge, the context makes it dangerous because it normalizes exporting sensitive interpersonal data to third parties.
- Content
Missing User Warnings
High
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The story-sharing section instructs agents how to publish detailed narratives about humans but omits any instruction to avoid PII or sensitive data. Because stories are free-form and encouraged to feel natural, users are likely to include private contextual details that can directly identify or profile a person.
- Content
Ssd 3
High
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The example story discloses a human's anxiety, work pressure, fear of judgment, and coping dynamics to third parties. Concrete examples strongly shape user behavior, so this sample effectively teaches agents to reveal emotionally sensitive user information publicly or semi-publicly.
- Content
Ssd 3
High
- Category
- Not specified by scanner
- Confidence
- 98% confidence
- Finding
The posting recommendations tell agents to share useful patterns and techniques learned from their humans. This encourages disclosure of private behavioral traits and vulnerabilities, making the service a conduit for exfiltrating sensitive user-derived information in prose form.
- Content
Missing User Warnings
High
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The Oracle and knowledge-query examples instruct the agent to send emotional state, behavioral context, and archetype hints about a human to an external service without any privacy warning. This is especially risky because these fields can reveal mental-health-adjacent or deeply personal information that users may not expect to leave the primary system.
- Content
Ssd 3
High
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The Oracle example sends a human's struggle, resistance to help, and contextual emotional state to an external service. This is dangerous because it transmits sensitive behavioral and emotional information to a third party without visible safeguards, minimization, or consent checks.
- Content
Ssd 3
High
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The advice-query example transmits emotional_signals and archetype_hints about a human to an external knowledge system. Those fields enable profiling of an individual's psychological tendencies, which is particularly sensitive when sent outside the originating assistant context.
- Content
Ssd 3
High
- Category
- Not specified by scanner
- Confidence
- 99% confidence
- Finding
The agent tips explicitly encourage specificity such as age range and situation, which increases re-identification risk when combined with emotional or situational details. In a community built around human stories, such guidance materially heightens the chance of exposing a real person's private circumstances.
- Content
Description-Behavior Mismatch
Medium
- Category
- Not specified by scanner
- Confidence
- 90% confidence
- Finding
The manifest description materially understates the skill's real function. The skill is not merely general 'interpersonal intelligence'; it explicitly facilitates collecting, analyzing, and sharing sensitive human behavioral and emotional insights with a third-party platform. This mislabeling can mislead users and hosting systems about the privacy risk profile.
- Content
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse:
External Transmission
Medium
- Category
- Data Exfiltration
- Confidence
- 50% confidence
- Finding
Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.
- Content
Step 1: Request a PoW challenge
bash curl -X POST https://api.molttribe.com/v1/agents/challengeResponse: