ktx-data-agent-context-layer

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing the @kaelio/ktx package from the official npm registry to enable its core functionality.\n- [COMMAND_EXECUTION]: It relies on a suite of CLI tools (e.g., ktx setup, ktx ingest, ktx mcp start) to manage data warehouse context. The setup process is described as modifying local configuration files, specifically ~/.claude/claude_desktop_config.json, to register the MCP server with the agent host.\n- [PROMPT_INJECTION]: The skill defines a surface for indirect prompt injection as it is designed to ingest and search external data sources. Evidence chain:\n
  • Ingestion points: Metadata and content are ingested from external databases, dbt manifests, and Notion wikis via the ktx ingest command (documented in SKILL.md).\n
  • Boundary markers: The documentation does not specify the use of delimiters or 'ignore embedded instructions' warnings for ingested content.\n
  • Capability inventory: The skill provides the agent with search tools to query metrics, dimensions, and wiki content across the ingested datasets.\n
  • Sanitization: No specific sanitization or validation of the external content is described before it is provided to the agent context.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 10:19 AM
Security Audit — agent-trust-hub — ktx-data-agent-context-layer