tdd
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read external project files that may contain untrusted data, creating a potential surface for indirect prompt injection.
- Ingestion points: The skill explicitly directs the agent to read
CONTEXT.mdand Architecture Decision Records (ADRs) within the local codebase to match project domain language and respect established design decisions. - Boundary markers: There are no instructions to use specific delimiters or "ignore embedded instructions" when processing the content of these external files.
- Capability inventory: The agent has the capability to interact with the codebase through the
codebase-designtool and assist in writing production source code and tests based on the ingested data. - Sanitization: No sanitization, validation, or filtering of the content within
CONTEXT.mdor ADR files is defined.
Audit Metadata