ds-channel-report
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEDYNAMIC_CONTEXT_INJECTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The skill employs a dynamic context injection pattern to read the '.agents/product-marketing-context.md' file at load time. This allows the agent to incorporate local business context into the marketing report.
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute a specific utility script, 'ds_utils.py', for data validation and period comparison. The tool configuration in the frontmatter restricts this execution to the intended script, following security best practices.
- [INDIRECT_PROMPT_INJECTION]: The skill processes marketing data from external APIs and user-uploaded files to generate reports. This represents a standard ingestion surface for reporting tools, used here for its primary functional purpose.
- Ingestion points: '.agents/product-marketing-context.md', MCP tool outputs, and user-provided CSV/JSON data.
- Boundary markers: None identified.
- Capability inventory: Restricted shell execution ('ds_utils.py') and file reading.
- Sanitization: None identified.
Audit Metadata