downloading-batch-export-files

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes raw PostHog data (events, persons, or sessions) which may contain attacker-controlled strings from external sources.
  • Ingestion points: Data is ingested into the agent context from the PostHog REST API at /api/projects/{project_id}/file_download_batch_exports/{run_id}/download/.
  • Boundary markers: The instructions do not define delimiters or specific markers to distinguish the raw data from potential instructions.
  • Capability inventory: The skill has capabilities to perform authenticated network requests and write files to the local system.
  • Sanitization: The content of the exported files (Parquet or JSONLines) is saved without sanitization or validation of the internal data fields.
  • [DATA_EXFILTRATION]: The skill performs network operations to download sensitive data and handles temporary signed access URLs.
  • The skill makes authenticated HTTP GET requests to PostHog's API infrastructure to fetch data exports.
  • The skill documentation correctly identifies a data exposure risk, noting that signed URLs grant temporary access to exported files and should not be shared with unrelated services.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 09:40 AM
Security Audit — agent-trust-hub — downloading-batch-export-files