k2f-case-study
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted user data to generate narrative content.
- Ingestion points: User-provided project notes, resume bullets, and outcome data processed during the content generation phase.
- Boundary markers: The instructions do not specify the use of delimiters or specific system instructions to differentiate user data from agent instructions.
- Capability inventory: The skill itself contains no scripts with file-system or network capabilities, but it utilizes the
k2fskill for final processing and export. - Sanitization: No sanitization or validation of the input text is mentioned.
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of a secondary component.
- Evidence: The workflow includes the command
npx skills add suzheng/k2f --skill k2fto install necessary formatting tools. - Context: The resource originates from the same author (suzheng) as the skill itself, representing a vendor-owned resource used for its intended primary purpose.
Audit Metadata