instantly-incident-runbook
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources such as webhook payloads, API response logs, and campaign metadata, which constitutes an indirect prompt injection attack surface.
- Ingestion points: Triage logic in
SKILL.mdinvolves analyzing campaign, account, and webhook failures from the Instantly API. - Boundary markers: The skill includes explicit instructions in the 'Output' and 'Instructions' sections to preserve redacted evidence and sensitive fields.
- Capability inventory: The skill is configured with
WebFetch,Write, andEdittools. - Sanitization: Security is addressed through a mandatory 'Approval Boundaries' section that forbids production mutations without explicit owner consent and requires diagnostic actions to remain read-only by default.
- [SAFE]: The skill uses official vendor-provided API endpoints (
api.instantly.ai) and references official platform packages. Credential safety is prioritized with instructions to use server-side secret management and avoid persisting or printing API keys.
Audit Metadata