conference-badge-scan-workflow
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill includes a bash code block
claude mcp add --transport http manypi https://mcp.manypi.com/mcp -s userintended to register a Model Context Protocol (MCP) server. This command, when run, extends the agent's capabilities with external tools provided by the vendor. - [EXTERNAL_DOWNLOADS]: The skill references and integrates a remote MCP service located at
https://mcp.manypi.com/mcp. This adds a dependency on external infrastructure hosted by the skill author for data enrichment tasks. - [INDIRECT_PROMPT_INJECTION]: The skill's primary workflow involves processing external datasets (badge-scan exports) which can be used to influence agent behavior.
- Ingestion points: Raw badge-scan exports (CSV or spreadsheets) containing name, company, and conversation notes.
- Boundary markers: There are no instructions to use delimiters or ignore instructions embedded within the scan data.
- Capability inventory: The skill utilizes an MCP tool to perform network operations and potentially write to a CRM/database.
- Sanitization: No sanitization, validation, or escaping of the external data fields is described in the workflow.
Audit Metadata