architect
Pass
Audited by Gen Agent Trust Hub on Jun 29, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (
gh) and Git for standard repository and issue management tasks, such as querying tickets, applying labels, and committing documentation to feature branches. These operations are aligned with the skill's stated purpose as a pipeline architect tool. - [PROMPT_INJECTION]: The skill processes untrusted data from GitHub issues and product documentation, constituting an Indirect Prompt Injection surface. It employs sanitization techniques, specifically normalizing idea slugs by lowercasing and removing special characters, to mitigate potential command injection or path traversal via user-controlled strings. Sub-agents are further constrained by prompts that restrict their output to specific architectural files.
- Ingestion points: GitHub issue body content and requirement documents in
docs/factorium/. - Boundary markers: Not explicitly present in sub-agent instructions.
- Capability inventory: File writing, directory creation, and version control operations.
- Sanitization: Derived idea slugs undergo character filtering (removing specials) and normalization before use in shell commands and file paths.
Audit Metadata