ktx-data-agent-context-layer
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs users to install the
@kaelio/ktxpackage from the NPM registry. This is a standard installation procedure for the core functionality of the context layer. - [INDIRECT_PROMPT_INJECTION]: The skill acts as an ingestion layer for external data sources including dbt models, Notion wikis, and database metadata. This creates an attack surface for indirect prompt injection if the ingested content contains instructions designed to manipulate the AI agent's behavior.
- Ingestion points: External data sources configured in
ktx.yaml(Notion, dbt, SQL warehouses). - Boundary markers: Not explicitly defined in the documentation.
- Capability inventory: Provides MCP tools for searching metrics and wiki content, and retrieving SQL definitions via CLI.
- Sanitization: The skill is documented as "read-only by design," which provides a significant safety barrier against data modification via injected instructions.
Audit Metadata