Comprehensive toolkit for generating best practice Helm charts and resources following current standards and conventions. Use this skill when creating new Helm charts, implementing Helm templates, or building Helm projects from scratch.
74
93%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
View guide
Low
Low-risk findings worth noting
The skill exposes the agent to untrusted, user-generated content from public third-party sources, creating a risk of indirect prompt injection. This includes browsing arbitrary URLs, reading social media posts or forum comments, and analyzing content from unknown websites.
The SKILL.md Chart Generation Workflow (Stage 2) explicitly requires a fallback "WebSearch" (e.g., searching "<operator> <CRD-kind> <version> kubernetes documentation spec") to fetch public third‑party documentation that the agent is expected to read and use to generate templates, so untrusted web content can directly influence its actions.
WebSearch fallback for Kubernetes CRD documentation
content-type · 2 sites
The SKILL.md explicitly instructs the agent to use WebSearch as a fallback to fetch public third-party Kubernetes documentation for CRD kinds, which the agent then reads and uses to generate Helm chart templates, creating an indirect prompt-injection vector.
SKILL.md
47
2. **Fallback to WebSearch:**
SKILL.md
49
"<operator>" "<CRD-kind>" "<version>" kubernetes documentation spec
context7 MCP
domain · 3 sites
The SKILL.md instructs the agent to use context7 MCP (a third-party Model Context Protocol service) to fetch CRD documentation, which is then used to generate Helm chart templates, creating an indirect prompt-injection vector.
SKILL.md
41
1. **Try context7 MCP first:**
SKILL.md
43
mcp__context7__resolve-library-id with operator name
SKILL.md
44
mcp__context7__get-library-docs with topic for CRD kind