summarize
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data from URLs, uploaded documents (PDF, Word, Markdown, text), and pasted content.
- Ingestion points: External content is introduced through the main summarization logic defined in
SKILL.mdvia the 'Take the source' section. - Boundary markers: The instructions do not define clear delimiters or specific instructions for the agent to ignore commands that may be embedded within the source documents.
- Capability inventory: The skill has the capability to write files to the
.overviews/directory in the repository root. - Sanitization: There are no explicit sanitization or validation steps for the content retrieved from remote or local sources before it is processed by the model.
Audit Metadata