airtable
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
curlfor API communication andpython3modules for JSON processing and URL encoding tasks.- [DATA_EXPOSURE_AND_EXFILTRATION]: Facilitates network communication with the official Airtable API domain (api.airtable.com). It correctly identifies the standard environment file for the agent platform as the appropriate location for storing API credentials.- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from Airtable records and table schemas, which constitutes an indirect prompt injection surface if the external data contains instructions for the agent. - Ingestion points: Retrieving record fields and table metadata from
https://api.airtable.com/v0. - Boundary markers: The instructions do not define specific delimiters to isolate external data from system instructions.
- Capability inventory: Ability to execute shell commands and network requests via
curlandpython3. - Sanitization: Uses
python3 urllib.parse.quoteto escape dynamic formula strings.
Audit Metadata