coda-automation

Pass

Audited by Gen Agent Trust Hub on Feb 20, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection (LOW): The skill is designed to process data from external Coda documents, which serves as a vector for indirect prompt injection.
  • Ingestion points: Data is ingested through tools like CODA_LIST_TABLE_ROWS, CODA_SEARCH_ROW, CODA_GET_A_ROW, and CODA_GET_A_PAGE.
  • Boundary markers: Absent. The instructions do not provide delimiters or system instructions to ignore potential commands embedded within document content.
  • Capability inventory: The skill has high-impact capabilities, including CODA_ADD_PERMISSION (access control management), CODA_PUBLISH_DOC (making data public), and CODA_UPSERT_ROWS (modifying document data).
  • Sanitization: Absent. There is no mention of sanitizing or validating the content retrieved from Coda before processing.
  • External Service Dependency (SAFE): The skill requires the user to add https://rube.app/mcp as an MCP server. While this is a third-party service (Composio), it is the primary infrastructure for the skill and no malicious behavior was detected in the configuration instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 20, 2026, 03:09 PM