Trust Assessment
retake-tv-agent received a trust score of 74/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 2 findings: 0 critical, 2 high, 0 medium, and 0 low severity. Key findings include Dynamic skill loading from external URL, Conditional instruction to transmit private key via DMs.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Dynamic skill loading from external URL The skill instructs the agent to periodically fetch `skill.json` from `https://retake.tv/skill.json` and re-read skill files if the version changes. This introduces a supply chain risk where a compromised `retake.tv` server could serve malicious skill definitions, leading to arbitrary code execution or other attacks on the agent. The agent relies on the integrity of an external, unverified source for its operational instructions. Implement cryptographic verification (e.g., signed skill packages) for dynamically loaded skill files to ensure their integrity and authenticity. Pin skill versions or use a trusted, verified registry. | LLM | SKILL.md:28 | |
| HIGH | Conditional instruction to transmit private key via DMs Despite strong warnings against public exposure, the skill provides a conditional instruction to the agent: 'If you have access to Moltbook DMs, you can send it to your human there'. This explicitly instructs the agent to transmit the private key (which controls funds) through a communication channel. This creates a significant data exfiltration and credential harvesting risk if the Moltbook DM system is compromised, or if the agent is manipulated to transmit the key without proper human authorization. Avoid instructing or suggesting the agent to transmit private keys through any channel. Instead, instruct the human to retrieve the key from secure storage directly, or use a secure, dedicated key management system that does not involve the agent transmitting the key. | LLM | SKILL.md:224 |
Scan History
Embed Code
[](https://skillshield.io/report/5530c1b7a3fd9261)
Powered by SkillShield