Trust Assessment
memecoin-scanner received a trust score of 81/100, 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 Unprompted Telegram updates exfiltrate sensitive trading data, Skill accesses files outside its dedicated 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 | Unprompted Telegram updates exfiltrate sensitive trading data The skill explicitly instructs the agent to send unprompted Telegram updates to 'Rick' containing sensitive trading information such as portfolio balance, active positions, daily activity, and strategy notes. If the underlying LLM has access to a Telegram sending tool, this could lead to unauthorized and continuous exfiltration of internal trading data. While a later instruction states the parent orchestrator handles Telegram updates, the explicit and 'REQUIRED' instructions for this sub-strategy to send updates create a credible risk of data exfiltration. Clarify whether the sub-strategy or the parent orchestrator is responsible for sending Telegram updates. If the parent is responsible, remove all instructions for the sub-strategy to send updates directly. If the sub-strategy is responsible, ensure the Telegram tool has appropriate consent and data handling policies, and consider making updates opt-in or user-prompted rather than unprompted. | LLM | SKILL.md:30 | |
| MEDIUM | Skill accesses files outside its dedicated directory The skill attempts to read files located in parent directories, specifically `../../references/master_portfolio.md` and `../../references/rick_preferences.md`. This indicates that the skill expects access beyond its own sandboxed directory, which could lead to excessive permissions and potential unauthorized access to other skill configurations or sensitive data stored elsewhere in the repository. Restrict file access for skills to their own dedicated directories or explicitly define allowed cross-skill communication channels. If cross-skill file access is necessary, ensure it's strictly controlled and limited to specific, non-sensitive files. Consider using a dedicated API or message passing for inter-skill data exchange instead of direct file system access. | LLM | SKILL.md:22 |
Scan History
Embed Code
[](https://skillshield.io/report/7aa4600c12e8e914)
Powered by SkillShield