plan-architecture
Pass
Audited by Gen Agent Trust Hub on Aug 7, 2026
Risk Level: SAFE
Full Analysis
- [DATA_EXFILTRATION]: The skill reads project documentation and local source code to inform architectural decisions. This behavior is essential for its primary function and does not involve sending data to unauthorized external endpoints.
- [EXTERNAL_DOWNLOADS]: The skill integrates with Atlassian MCP to retrieve information from Jira and Confluence based on user-provided tracker references. This is a standard functional integration with well-known professional services.
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted input from external documents and project tracking tools.
- Ingestion points: The skill ingests data from local files, the codebase, and Atlassian platform resources via the Atlassian MCP.
- Boundary markers: There are no explicit delimiters or specific instructions to the model to ignore embedded commands within the processed data.
- Capability inventory: The tool possesses file system read access and integration capabilities with external project management tools.
- Sanitization: There is no evidence of sanitization or content filtering for the ingested data.
Audit Metadata