skill-design-lineage
Pass
Audited by Gen Agent Trust Hub on Sep 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides shell script templates that use commands such as
git,mkdir,cat, andgrepto manage design artifacts and integrate with the local development environment. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: The skill reads previously saved design documents from the
~/.claude-octopus/designs/directory to provide context for downstream workflows like implementation and review (referenced inSKILL.md). - Boundary markers: There are no instructions to use delimiters or specific markers to isolate the content of loaded design documents from the agent's system prompt or other instructions.
- Capability inventory: The agent is granted capabilities to write to the file system, read directory contents, search file text, and execute git commands.
- Sanitization: The skill does not implement sanitization or validation mechanisms for the content of design documents before they are interpolated into the agent's context during cross-session discovery.
Audit Metadata