agent-data-cli

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute the agent-data CLI package and additional skills from the agent-data/skills GitHub repository. These resources belong to the official accounts of the skill's author.
  • [REMOTE_CODE_EXECUTION]: Employs npx to dynamically download and run the latest version of the agent-data CLI and installation scripts (e.g., init). This allows for runtime setup and updates of the vendor's tooling.
  • [COMMAND_EXECUTION]: The skill executes various shell commands using agent-data, npx, and skills to manage account setup, CLI installation, and API interactions. It uses flags like --yes to automate setup tasks.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes output from external APIs via agent-data call, creating a surface where third-party content could influence agent behavior if the source data contains malicious instructions.
  • Ingestion points: Output from agent-data call and agent-data docs (SKILL.md).
  • Boundary markers: None specified for external data output.
  • Capability inventory: Shell command execution via agent-data and npx (SKILL.md).
  • Sanitization: No specific filtering or sanitization of API response content is described before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 07:59 PM