coda-automation

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to add https://rube.app/mcp as an MCP server endpoint in their client configuration and references https://composio.dev for toolkit documentation and services. These are well-known services related to the skill's core functionality.
  • [COMMAND_EXECUTION]: The skill provides tools such as CODA_PUSH_A_BUTTON and CODA_UPSERT_ROWS that execute actions and data modifications within the Coda environment based on provided parameters.
  • [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests data from external Coda documents (e.g., through CODA_LIST_TABLE_ROWS and CODA_SEARCH_ROW).
  • Ingestion points: Data is retrieved from Coda documents, pages, and tables using tools like CODA_LIST_TABLE_ROWS and CODA_GET_A_ROW (SKILL.md).
  • Boundary markers: None identified in the skill instructions to delimit untrusted data from document content.
  • Capability inventory: The skill possesses write capabilities including updating rows (CODA_UPSERT_ROWS), managing document permissions (CODA_ADD_PERMISSION), and publishing documents (CODA_PUBLISH_DOC) (SKILL.md).
  • Sanitization: No specific sanitization or filtering of external document content is described.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 09:47 PM