keris-academic-search
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes
npxto fetch and execute instructions and helper utilities from the@nomadamas/k-skillpackage on the NPM registry. This is a functional requirement of the framework for managing updates and instruction sets.- [COMMAND_EXECUTION]: The instructions and Python scripts involve executing shell commands vianpxand running local scripts to interface with the RISS API. These operations are scoped to the skill's primary function of academic metadata retrieval.- [INDIRECT_PROMPT_INJECTION]: The skill processes XML metadata from the external RISS API to display search results, creating a potential surface for indirect prompt injection from upstream data. - Ingestion points: Search results are retrieved from
https://www.riss.kr/openApiwithinscripts/keris_academic.py. - Boundary markers: Results are formatted as structured JSON or clearly labeled text lists for the agent.
- Capability inventory: The skill is restricted to searching and reading academic metadata; it does not have write access or sensitive system capabilities.
- Sanitization: The implementation uses standard JSON encoding for structured output, providing basic protection against data formatting issues when results are presented to the agent.
Audit Metadata