k8s-generator
Pass
Audited by Gen Agent Trust Hub on Apr 6, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The skill instructs the agent to access sensitive environment configuration files. Evidence: In Phase 1 of SKILL.md, the agent is directed to use Glob and Grep to locate and read .env files to extract environment variables for manifest generation.
- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection via the processing of untrusted project files. Ingestion points: Reads Dockerfile, .env, and other application-specific files during analysis as described in SKILL.md. Boundary markers: No delimiters or specific instructions to ignore malicious commands embedded in these external files are provided. Capability inventory: The skill allows the agent to use the Write tool to generate files based on the content of these untrusted inputs. Sanitization: There is no validation or escaping of the content retrieved from application files before it is processed.
Audit Metadata