migrating-from-customer-io
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is designed for a legitimate migration task between two well-known services (Customer.io and PostHog).
- [SAFE]: Explicit privacy constraints are enforced. The instructions strictly forbid fetching or passing sensitive "people data," including customer profiles, segment members, or delivery event data, to minimize privacy risks.
- [SAFE]: The skill implements a "human-in-the-loop" design, requiring the agent to perform dry-runs (
dry_run: true), show previews to the user, and wait for explicit sign-off before importing or enabling workflows. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests untrusted data from an external API (Customer.io).
- Ingestion points: Reads campaign data, actions, templates, and snippets via the
cio_read_apitool (SKILL.md). - Boundary markers: None identified between external data and the agent's context.
- Capability inventory: Includes tool calls for
project-get,switch-project, andworkflows-import-customerio(SKILL.md). - Sanitization: None specified for the content retrieved from Customer.io.
- Assessment: The risk is significantly mitigated by the skill's narrow scope, the requirement for human verification of all drafts, and the explicit prohibition of transferring PII (Personally Identifiable Information).
Audit Metadata