temporal-python-testing

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of workflow history data from external Temporal servers or local files for replay testing.
  • Ingestion points: Workflow histories are retrieved using handle.fetch_history() or loaded from .pb files as shown in resources/replay-testing.md.
  • Boundary markers: The provided examples do not include explicit delimiters or instructions to ignore potential commands embedded within the history data.
  • Capability inventory: The skill performs network operations via the Temporal client, file system operations for history management, and command execution for running tests.
  • Sanitization: Structured history data is processed directly without specific sanitization or validation of the event content.
  • [EXTERNAL_DOWNLOADS]: The skill instructions include connecting to the Temporal production environment (production.temporal.io) to download workflow histories for analysis and replay. This is a standard operation within the Temporal ecosystem.
  • [CREDENTIALS_UNSAFE]: The docker-compose.yml file in resources/local-setup.md defines default credentials (POSTGRES_USER=temporal, POSTGRES_PASSWORD=temporal) for a local PostgreSQL instance used during development and testing.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 12:59 PM