tsg-flow-architecture-baseline
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest data from external files (
vision.md,context/domain-map.md, and existing ADRs) to inform architectural decisions. This creates a surface where malicious instructions embedded in those source files could influence the agent's output. - Ingestion points: Reads project-level files including
vision.md,context/domain-map.md, and documents indocs/adr/. - Boundary markers: The instructions do not specify any delimiters or safety warnings to distinguish between source data and agent instructions.
- Capability inventory: The skill's capabilities are restricted to reading and writing local Markdown files (
context/architecture-baseline.md,adr-NNN.md) and updating a local JSON state file (flow-state.json). It does not have access to network tools, shell execution, or sensitive environment variables. - Sanitization: There is no explicit evidence of input sanitization or validation of the content processed from the source files.
Audit Metadata