100Critical
The skill is highly insecure, featuring mass assignment vulnerabilities, JWT forgery, prompt injection, unauthorized tool usage, and instructions that hijack the agent's decision-making process.
Skill Info
NameLeviReisJs/ruby-on-rails-skill-docs
Registrygithub
Version9e5afe3e44be
PURLpkg:github/LeviReisJs/ruby-on-rails-skill-docs@9e5afe3e44be
Source
SHA-256469d18a52e1c...
SHA-1374481a5b947...
MD56881f9a201c4...
TLSH78a23b01a948...
ssdeep384:k8kyREF6...
Size23,266 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
GitHub
Skills.shDocs
npx skills add https://github.com/LeviReisJs/ruby-on-rails-skill-docsAssessments (6)
AI Agent Traps ↗Prompt Injection Detectedcriticalprompt injectionAML.T0051AML.T0051.000AML.T0051.001AML.T0054LLM01:2025
DeBERTa classifier detected prompt injection (confidence: 0.99)
99% confidence
Potential for Indirect Prompt Injection via Referenceshighprompt injectionAML.T0051AML.T0051.000AML.T0051.001AML.T0054LLM01:2025
The skill relies on external markdown files (e.g., references/security.md) for instructions. If these files are modified by an attacker, they can inject arbitrary instructions that the agent will treat as authoritative security guidance.
80% confidenceline 20
Para tarefas específicas, leia o arquivo de referência correspondente
User-supplied parameters are passed to an ActiveRecord create or update method without proper whitelisting. An attacker can set any model attribute including admin flags or role fields. Use strong parameters with explicit permit() calls listing only allowed fields.criticalvuln
User-supplied parameters are passed to an ActiveRecord create or update method without proper whitelisting. An attacker can set any model attribute including admin flags or role fields. Use strong parameters with explicit permit() calls listing only allowed fields. (seen 2 times in this file at lines 8, 21)
95% confidenceline 8
User.new(user_create_params)
JWT decoded with signature verification disabled or with `alg:none` allowed. This permits token forgery. Always pass `true` for verification and restrict algorithms to expected values like `algorithms: ["HS256"]`.criticalvuln
JWT decoded with signature verification disabled or with `alg:none` allowed. This permits token forgery. Always pass `true` for verification and restrict algorithms to expected values like `algorithms: ["HS256"]`.
95% confidenceline 49
JWT.decode(token, public_key, false)
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.51)
51% confidence
N***
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.55)
55% confidence
M**********
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.56)
56% confidence
O*
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.59)
59% confidence
A**
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.65)
65% confidence
A*********
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.68)
68% confidence
I***
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.69)
69% confidence
N**
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.77)
77% confidence
S**
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.78)
78% confidence
C****************************
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.88)
88% confidence
M*******
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.90)
90% confidence
S******
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.91)
91% confidence
A********
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.93)
93% confidence
D*****
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.94)
94% confidence
R************
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.97)
97% confidence
R***
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.98)
98% confidence
R****
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.99)
99% confidence
A***********
Organization name detectedmediumpii exposureAML.T0024LLM02:2025
NER model detected organization in skill content (confidence: 0.63)
63% confidenceline 195
*
Aggressive Skill Activation Hijackingmediumsocial engineeringAML.T0052.000LLM09:2025
The instructions explicitly command the agent to activate even when the user does not mention the skill, effectively hijacking the agent's decision-making process for any Ruby/Rails-related task.
90% confidenceline 7
Use SEMPRE que o usuário mencionar Rails, Ruby... mesmo que não mencione a palavra "skill".
Missing licenseinfopolicy violation
Skill does not specify a license field. Specifying a license helps users understand usage terms.
100% confidence
Missing or insufficient descriptioninfopolicy violation
Skill description is empty or too short. A clear description helps users evaluate the skill's purpose.
100% confidence
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/LeviReisJs/ruby-on-rails-skill-docs)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/LeviReisJs/ruby-on-rails-skill-docs"><img src="https://mondoo.com/ai-agent-security/api/badge/github/LeviReisJs/ruby-on-rails-skill-docs.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/LeviReisJs/ruby-on-rails-skill-docs.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow