downloading-batch-export-files
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements data export workflows using official PostHog MCP tools and REST API endpoints. All operations are performed within the existing PostHog authentication context.
- [SAFE]: Data protection measures are explicitly included, such as advising the agent to save binary or large JSONLines data directly to files rather than outputting contents to the chat interface. It also includes a specific warning against sharing temporary signed URLs with unrelated services.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface where user-provided input is used to construct data queries.
- Ingestion points: User-provided HogQL queries, event filters (
include/exclude), and data intervals inSKILL.md. - Boundary markers: None specified; instructions do not define delimiters for user-supplied HogQL strings.
- Capability inventory: The skill uses MCP tools for API interaction, performs network requests for file downloads, and performs local file writes.
- Sanitization: The skill instructs the agent to clarify user inputs, use specific ISO 8601 formats, and prefer limiting data exports to the minimum necessary to fulfill the request.
Audit Metadata