skills/vm0-ai/vm0-skills/plain/Gen Agent Trust Hub

plain

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: Uses system commands like curl and jq to communicate with the Plain.com API and parse results. GraphQL requests are constructed by writing JSON payloads to /tmp/plain_request.json before execution.
  • [EXTERNAL_DOWNLOADS]: Performs network operations to core-api.uk.plain.com and plain.com. These are the official domains for the Plain.com service and are used for legitimate API interactions and documentation access.
  • [PROMPT_INJECTION]: 1. Ingestion points: The skill retrieves data from an external source (Plain API), including customer-generated support thread content and user details (SKILL.md). 2. Boundary markers: Absent. There are no instructions for the agent to use delimiters or ignore instructions embedded within the customer data. 3. Capability inventory: The skill has the ability to write to the local file system (/tmp) and perform outbound network requests via curl (SKILL.md). 4. Sanitization: Not specified. The agent is not instructed to sanitize or validate the content retrieved from threads before processing it.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 04:44 PM