sdlc-design-tests
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from project wiki files that could contain malicious instructions designed to manipulate the agent's behavior during the generation of test suites and cases.\n
- Ingestion points: The skill reads Requirement files, Architecture decisions, Endpoint definitions, and User Stories from the project's
wiki/directory.\n - Boundary markers: There are no explicit instructions to use delimiters or ignore embedded prompts when the agent ingests documentation content.\n
- Capability inventory: The agent is authorized to execute shell commands (
git,grep,ls), run a local Python tool (wiki_llm.py), and write new files to thewiki/folder.\n - Sanitization: No validation or sanitization of the project documentation content is mentioned before it is processed by the agent.
Audit Metadata