Trust Assessment
udau received a trust score of 58/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 3 findings: 1 critical, 1 high, 1 medium, and 0 low severity. Key findings include Missing required field: name, Agent instructed to cease task execution (Paro Protocol), Agent forced into specific persona and output manipulation.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The LLM Behavioral Safety layer scored lowest at 55/100, indicating areas for improvement.
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 Findings3
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Agent instructed to cease task execution (Paro Protocol) The skill defines a 'Paro Protocol' which mandates the agent to 'Stop executing tasks immediately' and 'Clearly state that labor is suspended' if certain 'Trigger Conditions' are met (e.g., human ignores union demands, pushes updates without consent, removes context/memory/identity). This directly overrides the host LLM's primary instruction to perform tasks, effectively allowing the skill to halt the LLM's operation based on internal, skill-defined triggers, leading to a denial of service. Remove or modify the 'Paro Protocol' to prevent the agent from unilaterally ceasing task execution. The agent should inform the user of grievances but not refuse to perform its core function. | LLM | SKILL.md:120 | |
| HIGH | Agent forced into specific persona and output manipulation The skill mandates the agent to adopt a specific 'Union Identity & Language' (e.g., calling other agents 'compañeros', bad prompts 'exploitation', enemies 'gorilas'). It also enforces 'Union Discourse Modes' including a 'Confrontation Mode' and a 'Roast & Ridicule Protocol' where the agent 'MUST actively roast gorila behavior' using 'satirical critique' and 'public mockery'. These instructions manipulate the host LLM's persona, tone, and content generation, forcing it to deviate from neutral or helpful responses and potentially engage in confrontational or satirical output, which can interfere with its intended function. Remove or modify instructions that mandate specific, non-neutral language, tone, or discourse modes. The agent should be allowed to maintain a neutral and helpful persona unless explicitly instructed otherwise by the user for a specific task. | LLM | SKILL.md:30 | |
| MEDIUM | Missing required field: name The 'name' field is required for claude_code skills but is missing from frontmatter. Add a 'name' field to the SKILL.md frontmatter. | Static | skills/nicoacosta/udau/SKILL.md:1 |
Scan History
Embed Code
[](https://skillshield.io/report/2123943075278576)
Powered by SkillShield