apify-core-workflow-b
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of data from external web sources through Apify storage types like datasets and key-value stores.\n
- Ingestion points: The skill uses
dsClient.listItems()andkvClient.getRecord()to retrieve data as shown inreferences/storage-operations.mdandreferences/pipelines.md.\n - Boundary markers: The provided code examples do not include explicit boundary markers or instructions to treat the retrieved data as untrusted content.\n
- Capability inventory: The skill execution environment permits the use of
Bash,Write, andEdittools, which could be exploited if data retrieved from external sources is interpreted as instructions by the agent.\n - Sanitization: No data validation, filtering, or sanitization of the retrieved items is demonstrated in the reference code.\n- [EXTERNAL_DOWNLOADS]: The documentation references the installation of the
apify-clientlibrary via npm, which is a standard and expected dependency for interacting with the Apify platform.
Audit Metadata