The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the configuration of an external MCP server located at https://rube.app/mcp. This third-party service mediates the connection between the agent and the Coda API.
[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it ingests untrusted data from external sources.
Ingestion points: Tools such as CODA_LIST_TABLE_ROWS, CODA_SEARCH_ROW, CODA_GET_A_PAGE, and CODA_GET_A_FORMULA read content directly from Coda documents and tables into the agent's context.
Boundary markers: There are no instructions or markers defined in SKILL.md to help the agent distinguish between its instructions and the data retrieved from Coda.
Capability inventory: The skill possesses significant capabilities, including modifying table data (CODA_UPSERT_ROWS), altering document permissions (CODA_ADD_PERMISSION), and publishing documents to the public web (CODA_PUBLISH_DOC).
Sanitization: The skill does not implement or recommend any sanitization, filtering, or validation of the data retrieved before it is processed by the agent.
[DATA_EXFILTRATION]: The skill contains tools that facilitate moving data outside of a private environment, specifically CODA_BEGIN_CONTENT_EXPORT (which generates temporary download links), CODA_ADD_PERMISSION (which can grant external access), and CODA_PUBLISH_DOC (which makes content publicly accessible via a URL).

coda-automation