architecture-advisor
Pass
Audited by Gen Agent Trust Hub on Oct 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill researches the existing codebase, reads configuration files, and reviews existing Architecture Decision Records (ADRs) to inform its advice. This ingestion of potentially untrusted data from the filesystem creates a surface for indirect prompt injection if the project files contain malicious instructions.
- Ingestion points: Local codebase files, system configuration files, and architectural documentation within the
docs/adr/directory. - Boundary markers: The skill does not define specific delimiters or "ignore previous instructions" warnings to isolate the content of the files it reads from its own operational logic.
- Capability inventory: The agent has the ability to read arbitrary files from the codebase and write new ADR documents to the filesystem.
- Sanitization: There is no evidence of content validation, escaping, or filtering of the information retrieved from the project files before it is processed by the agent.
Audit Metadata