find-journalists
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Bash to execute newsjack CLI commands for news searching, journalist enrichment, and credit balance checks as part of its core workflow.
- [EXTERNAL_DOWNLOADS]: The skill retrieves information from external news articles and uses the Medialyst API via CLI or MCP tools to fetch journalist data and metadata.
- [DATA_EXFILTRATION]: As part of its intended function, the skill transmits user-provided article URLs and pitch descriptions to the Medialyst API to perform journalist enrichment.
- [PROMPT_INJECTION]: The skill processes untrusted content from external news results and article URLs (ingestion point: newsjack news search, journalists enrich). It lacks explicit boundary delimiters in its prompt logic but references external guidelines (skills/ETHICS.md, skills/WHY-NOT-SPAM.md). Its capability inventory includes Bash execution and MCP tool usage. The skill provides remediation instructions for defensive JSON parsing and literal command string scanning to mitigate risks from malicious data ingestion.
Audit Metadata