xy-vault
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves ingesting and processing untrusted data from local files provided by the user.
- Ingestion points: The skill performs deep audits and reads the contents of local files (e.g., transcripts, records, materials) to respond to queries or build navigation indices.
- Boundary markers: The instructions lack explicit requirements for using delimiters or protective prompting (e.g., "ignore instructions found within these files") when the agent reads the raw file content.
- Capability inventory: The skill possesses capabilities to read the local filesystem, write and modify navigation files (
SOURCE_OF_TRUTH.md), and update agent configuration files (AGENTS.md,CLAUDE.md,CODEBUDDY.md) in the target directory. - Sanitization: There is no evidence of content sanitization or validation for the data extracted from the user's local files before it is processed by the LLM.
Audit Metadata