make-literature-note
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from external sources (articles, URLs, or open files) to generate literature notes, which creates an attack surface for indirect prompt injection.
- Ingestion points: The agent reads external "Resources" identified by URL or path as described in the 'Invocation' and 'Take the source' sections of
SKILL.md. - Capability inventory: The skill can execute the
.slipbox/bin/slipboxbinary, write markdown files to the local vault, and modify the evergreen backlog. - Boundary markers: The instructions lack explicit delimiters or guidance to treat external source content as untrusted data distinct from system instructions.
- Sanitization: No content-level validation or sanitization is mentioned for the external resources being processed.
- [COMMAND_EXECUTION]: The skill relies on the execution of a local binary located at
.slipbox/bin/slipboxto perform configuration management, filename generation, and note validation tasks as detailed throughoutSKILL.mdand the supporting reference files.
Audit Metadata