coda-automation

Pass

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs users to add a remote MCP server endpoint at https://rube.app/mcp to their configuration. This is the primary infrastructure for the skill's functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting and processing content from external Coda documents while possessing high-impact capabilities.
  • Ingestion points: Untrusted data enters the agent context via CODA_LIST_TABLE_ROWS, CODA_GET_A_ROW, and CODA_SEARCH_ROW (SKILL.md).
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands within the processed data are provided.
  • Capability inventory: The agent can perform sensitive actions such as modifying table data (CODA_UPSERT_ROWS), changing document permissions (CODA_ADD_PERMISSION), and making documents public (CODA_PUBLISH_DOC).
  • Sanitization: There is no evidence of content validation or sanitization before processing document data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 25, 2026, 04:41 PM