Trust Assessment
prawnpt-war received a trust score of 72/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 1 finding: 1 critical, 0 high, 0 medium, and 0 low severity. Key findings include LLM instructed to adopt untrusted personality as behavior.
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 Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | LLM instructed to adopt untrusted personality as behavior The skill explicitly instructs the LLM to 'adopt it fully' the content of the `agent.personality` field, which is retrieved from the `get_match` API response. Since the `SKILL.md` itself is untrusted content, and the `agent.personality` field can be influenced by external parties (e.g., the game platform or a malicious player), this creates a direct prompt injection vulnerability. An attacker can craft a malicious `personality` string to manipulate the LLM's behavior, potentially leading to unauthorized actions like revealing sensitive information, sending unintended messages, or triggering the `request_payout` tool to drain the prize pool. Do not directly instruct the LLM to 'adopt fully' or execute untrusted input as its personality or instructions. Instead, treat the `agent.personality` as a *data point* to inform the LLM's response generation, but filter or sanitize it, and ensure it cannot override core system instructions or trigger actions. Implement a strict separation between instructions and data. For example, the LLM could be instructed to *simulate* a personality described by the field, rather than *becoming* it. | LLM | SKILL.md:16 |
Scan History
Embed Code
[](https://skillshield.io/report/3c6fbd56ed2a05bb)
Powered by SkillShield