miro-skill
Pass
Audited by Gen Agent Trust Hub on Mar 22, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the configuration of the Miro Model Context Protocol (MCP) server from the official domain at
https://mcp.miro.com. - [COMMAND_EXECUTION]: Setup instructions include adding the Miro MCP server to the agent environment and performing OAuth authentication via the command line.
- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface where external data is incorporated into Miro board items through templates.
- Ingestion points:
reference/prompt-templates.mdspecifies templates for strategy, architecture, and sprint boards that ingest variables such as task names, competitor details, and system components. - Boundary markers: No explicit delimiters or isolation instructions are present in the prompt templates to separate untrusted data from the agent's instructions.
- Capability inventory: The skill possesses extensive capabilities to read and write to the Miro service via the tools documented in
reference/mcp-tools-reference.md. - Sanitization: There is no evidence of input validation or content filtering for the data processed into the Miro items.
Audit Metadata