domain-glossary
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external, potentially untrusted content, which creates a surface for indirect prompt injection.
- Ingestion points: According to step 1 of the Process section in SKILL.md, the skill reads from the current conversation context, attached documents, and Project Requirements Documents (PRDs).
- Boundary markers: There are no instructions defining boundary markers or explicit delimiters to isolate the source text, nor are there instructions for the agent to ignore potentially malicious embedded commands within those sources.
- Capability inventory: The skill utilizes file-writing capabilities to create and update a specific local file,
DOMAIN_GLOSSARY.md, in the working directory. - Sanitization: The process lacks specific sanitization, validation, or filtering steps for the content extracted from external documents before it is merged into the glossary file.
Audit Metadata