exa-get-contents

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION] (MEDIUM): The skill is susceptible to Indirect Prompt Injection (Category 8). It ingests untrusted data from external URLs which are then processed for summarization and structured extraction.
  • Ingestion points: Data enters the agent's context through the output of exa-ai get-contents <URL>.
  • Boundary markers: No specific delimiters or instructions are provided to the agent to treat the fetched content as untrusted or to ignore instructions embedded within it.
  • Capability inventory: The skill executes shell commands (exa-ai, jq) and processes the resulting text for decision-making.
  • Sanitization: There is no evidence of sanitization or filtering of the content retrieved from external websites.
  • [COMMAND_EXECUTION] (LOW): The skill requires the execution of the exa-ai CLI and jq utility. While these are standard tools for this skill's purpose, they involve direct interaction with the host's shell environment.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 12:50 PM