jq

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The documentation includes bash examples for shell pipelines using commands such as cat, sort, uniq, and xargs to manipulate JSON data streams.- [EXTERNAL_DOWNLOADS]: Demonstrates fetching data from well-known sources like the GitHub API using curl.- [CREDENTIALS_UNSAFE]: Includes command examples that read the agent's internal skill configuration file located at ~/.agents/.skill-lock.json.- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection when processing external data.
  • Ingestion points: JSON data from local files (data.json, users.json) and remote API responses (api.github.com).
  • Boundary markers: No delimiters or instructions to ignore embedded commands are present in the examples.
  • Capability inventory: The skill demonstrates capabilities for subprocess execution (bash) and network operations (curl).
  • Sanitization: Data is processed directly without escaping or validation before being passed to the shell.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 11:39 PM