teleport
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill functions by processing user-supplied descriptions of source and destination environments, which presents a surface for indirect prompt injection.
- Ingestion points: User-provided descriptions of systems, services, and environments used to trigger the migration spell (SKILL.md).
- Boundary markers: The procedure instructs the agent to "Restate the target, the success condition, and any no-touch boundaries" and to "Define boundaries" (SKILL.md).
- Capability inventory: The skill is tagged with actuation, execution, and automation, and the instructions describe high-privilege operations including data replication, auth migration, and traffic routing.
- Sanitization: The instructions do not specify any validation, escaping, or filtering of the external environment descriptions before they are used to generate runbooks and execution sequences.
- [NO_CODE]: The skill consists entirely of natural language instructions in markdown format and does not include any scripts, compiled binaries, or external runtime dependencies.
Audit Metadata