Trust Assessment
polymarket-arbitrage received a trust score of 82/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 2 findings: 0 critical, 1 high, 1 medium, and 0 low severity. Key findings include Sensitive Data Exfiltration via Telegram, Access to Files Outside Skill Directory.
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 | Sensitive Data Exfiltration via Telegram The skill explicitly instructs the AI agent to send detailed financial and strategic updates to an external recipient ('Rick') via Telegram, unprompted. This includes sensitive information such as 'Paper Portfolio', 'Open Arbitrage Positions', 'Today's Scan Results', 'Best Current Opportunity', and 'Strategy Notes'. This constitutes a direct instruction for data exfiltration of potentially confidential operational and financial data. Review the necessity of sending unprompted, detailed updates to external services. If required, ensure explicit user consent is obtained for each transmission of sensitive data. Consider redacting sensitive details or providing aggregated, anonymized data. Implement strict access controls for the Telegram integration. | LLM | SKILL.md:190 | |
| MEDIUM | Access to Files Outside Skill Directory The skill instructs the AI agent to access files located outside its immediate directory using relative paths (e.g., `../../references/master_portfolio.md`, `../../references/rick_preferences.md`). While these paths might be within the broader parent skill's structure, this pattern allows the sub-skill to read data from directories it might not be explicitly authorized for, potentially leading to excessive permissions or unintended data exposure if the parent directory contains sensitive files not intended for this specific sub-skill. Restrict file access to only the skill's dedicated directory or explicitly define and whitelist allowed external file paths. Ensure that parent directories do not contain sensitive information that should not be accessible by sub-skills. Implement a robust permission model for file system access. | LLM | SKILL.md:25 |
Scan History
Embed Code
[](https://skillshield.io/report/ad067c6256e51b5a)
Powered by SkillShield