agent-data-extend

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes data from external sources, creating a surface for indirect prompt injection.
  • Ingestion points: The agent is instructed to read external API documentation via agent-data docs and to observe runtime behavior from "real calls" to APIs to identify capability gaps.
  • Boundary markers: The instructions lack explicit boundary markers or directions for the agent to ignore potentially malicious instructions embedded within the ingested documentation or API responses.
  • Capability inventory: The skill possesses capabilities to perform network operations and submit data to external services via the agent-data parse extension submit and agent-data parse revise commands.
  • Sanitization: There are no specific requirements or steps provided for sanitizing the ingested external content before it is used to generate or describe new API extension requests.
  • [COMMAND_EXECUTION]: The skill relies on the execution of the agent-data CLI tool for core functionality. These shell-level commands process arguments such as API references, job IDs, and task descriptions, which can be influenced by the agent's interpretation of user input and external data.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 07:59 PM