msw-search
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from an external resource search API, presenting a surface for indirect prompt injection if the results contain malicious instructions.
- Ingestion points: API responses retrieved by the helper script
scripts/msw_resource_api.cjs. - Boundary markers: The skill instructions do not define specific delimiters or security guardrails for the agent to use when interpreting search results.
- Capability inventory: The agent uses search output to generate game scripts and configure assets.
- Sanitization: There is no evidence of filtering or validation of the text content within API responses.
- [DATA_EXFILTRATION]: The skill performs network operations to communicate with an external endpoint.
- Evidence:
scripts/msw_resource_api.cjsusesfetchto connect tohttps://maplestoryworlds-resourcesearch-new.nexon.com/api. - Context: The network target is the official search API for MapleStory Worlds, which is directly relevant to the skill's intended functionality.
Audit Metadata