graphql-schema-design
Pass
Audited by Gen Agent Trust Hub on Jul 8, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect prompt injection surface identified in the Change Review mode.
- Ingestion points: The skill ingests external file content by prompting the agent to run 'git diff main' in Mode 2 of SKILL.md.
- Boundary markers: No delimiters or specific instructions are provided to the agent to isolate the ingested code content from its instructions, meaning malicious content within a code diff could potentially influence the agent's reasoning.
- Capability inventory: The skill executes git commands and performs architectural analysis, but it is explicitly constrained from writing implementation code in its system instructions.
- Sanitization: No sanitization or validation is applied to the data retrieved from the git diff before it is processed by the agent.
Audit Metadata