ddd-context
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to create directory structures and manage files based on user input. - [INDIRECT_PROMPT_INJECTION]: The skill uses user-supplied arguments to construct filesystem paths, creating a potential surface for path injection.
- Ingestion points: The
$ARGUMENTSvariable is parsed for the context name and used directly in file and directory creation steps inSKILL.md. - Boundary markers: The skill includes an explicit instruction for the agent to validate that the name is in kebab-case and does not already exist.
- Capability inventory: The skill has access to
Bash,Write,Edit, and several vendor-specific storage tools to modify the environment. - Sanitization: There is no formal escaping or sanitization defined for the context name beyond the instructional validation step.
Audit Metadata