prisma-next-debug
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and interpret structured error envelopes (JSON/TS objects) provided by users to diagnose framework failures. This creates a potential surface where malicious instructions could be embedded within error fields like
summary,why, ormetato influence the agent's next routing decision. - Ingestion points: User-pasted error envelopes from CLI output or runtime exceptions, as described in the "Key Concepts" section of
SKILL.md. - Boundary markers: The instructions do not define specific delimiters or "ignore embedded instructions" warnings for the agent when parsing the envelope content.
- Capability inventory: The skill instructs the agent to recommend user actions such as
prisma-next init,db update, ordb sign, and routes the workflow to other sibling skills (e.g.,prisma-next-migrations) which may have more extensive file-system or database access. - Sanitization: No explicit sanitization or validation of the text content within the error envelope fields is described.
Audit Metadata