architect
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests and processes content from the local codebase and documentation.
- Ingestion points: The skill reads
docs/CONTEXT_MAP.md,docs/contexts/*/CONTEXT.md,docs/UBIQUITOUS_LANGUAGE.md, and performs general codebase exploration as part of its discovery process. - Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from executing instructions that might be embedded within the project files it analyzes.
- Capability inventory: The skill uses tools to read files, explore directories, and write output to
docs/contexts/<name>/specs/ordocs/contexts/<name>/adrs/based on architectural decisions. - Sanitization: No validation or sanitization mechanisms are described for the content retrieved from the codebase before it is used in the sub-agent design process or comparison phase.
Audit Metadata