dayz-persistence

Pass

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill provides instructions for the agent to ingest and process untrusted data from external sources, which could potentially contain malicious instructions intended to influence agent behavior.
  • Ingestion points: The agent is instructed to read data using JsonFileLoader<T>.LoadFile (SKILL.md), ParamsReadContext.Read (references/vanilla-stream.md), and FileSerializer.Read (references/sidecar-files.md).
  • Boundary markers: The skill advocates for "fail-closed" logic, requiring explicit version checks and header validation to identify and isolate data (references/migration-matrix.md).
  • Capability inventory: The instructions involve file system operations such as DeleteFile, CopyFile, and FPrint, which could be abused if an injection is successful (references/sidecar-files.md).
  • Sanitization: The skill mandates the use of structured data (JSON) with schema validation and checking the success of all read operations before applying state mutations (references/sidecar-files.md, references/vanilla-stream.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 5, 2026, 12:31 AM