testing-strategies
Warn
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill contains non-printable control characters and null bytes (such as \u0000, \u000c, \u0010, \u0002, \u0014, and \u0018) within its ASCII diagrams. These characters can be used in obfuscation techniques to hide content from plain-text scanners or to bypass safety filters.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through the ingestion of untrusted variable data.\n
- Ingestion points: The frontmatter defines 'language', 'test_type', and 'methodology' variables that are processed at runtime (SKILL.md).\n
- Boundary markers: No specific delimiters or safety instructions are provided to isolate user-controlled variables from the rest of the prompt logic.\n
- Capability inventory: The skill is configured with 'Bash' and 'Edit' tools, allowing for command execution and file modification (SKILL.md).\n
- Sanitization: While inputs are restricted by enums, the lack of boundary isolation remains a structural risk for complex injection patterns.
Audit Metadata