skills/x-cmd/skill/jq/Gen Agent Trust Hub

jq

Pass

Audited by Gen Agent Trust Hub on Apr 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the x jq CLI tool to process, filter, and transform JSON data. It also uses command substitution $(date) in examples to pass dynamic arguments to the tool. Evidence found in SKILL.md and SKILL.zh.md.\n- [EXTERNAL_DOWNLOADS]: The skill documentation indicates that x jq automatically downloads the jq binary if it is not present on the system. This is a documented feature of the x-cmd ecosystem and is treated as a vendor-provided resource.\n- [PROMPT_INJECTION]: The skill processes untrusted JSON data from external files or standard input, which serves as a potential vector for indirect prompt injection.\n
  • Ingestion points: JSON files (e.g., package.json, data.json) and piped input mentioned in SKILL.md and SKILL.zh.md.\n
  • Boundary markers: None identified in the prompt instructions to delineate data from instructions.\n
  • Capability inventory: Execution of the x jq utility for data transformation.\n
  • Sanitization: No explicit sanitization or escaping of the processed data is mentioned to prevent the agent from interpreting content as instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 10, 2026, 05:40 AM