reasoning
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external documentation (via Context7 MCP) and user-provided tasks to ground its reasoning process. This introduces a potential surface for indirect prompt injection if retrieved documentation contains adversarial instructions. This is a common architectural risk for skills that incorporate external data for grounding.
- [EXTERNAL_DOWNLOADS]: The skill systematically retrieves technical documentation from Context7. This is a functional requirement for grounding reasoning in up-to-date library information and is performed through standard Model Context Protocol (MCP) interfaces.
- [COMMAND_EXECUTION]: The skill references the use of tool commands, such as
/jira:reason, to integrate its reasoning frameworks with external issue tracking systems. This is consistent with its stated purpose of analysis and debugging within a managed environment.
Audit Metadata