jq-yq

Pass

Audited by Gen Agent Trust Hub on Feb 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • EXTERNAL_DOWNLOADS (LOW): Reference guides suggest installing tools via system package managers (apt, brew, choco) and Python's pip. While these are trusted sources, they represent external dependencies.
  • COMMAND_EXECUTION (LOW): The skill documentation and provided shell scripts (scripts/json-parser.sh, scripts/yaml-parser.sh) utilize CLI tools like jq, yq, and kubectl. These are used according to their primary purpose of data processing and infrastructure management.
  • DATA_EXFILTRATION (LOW): Examples demonstrate fetching data from external URLs using curl. While these are documentation placeholders, users should ensure they do not pipe sensitive information to untrusted external endpoints.
  • PROMPT_INJECTION (LOW): The skill is susceptible to Indirect Prompt Injection (Category 8) as its primary function is to ingest and transform untrusted structured data.
  • Ingestion points: Network responses from curl, cluster status from kubectl, and local configuration files.
  • Boundary markers: None found; the provided scripts do not use specific delimiters to isolate untrusted data.
  • Capability inventory: High, as the processed data is often used to drive further command-line operations or system configurations.
  • Sanitization: No sanitization or schema validation is implemented in the provided shell script examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 21, 2026, 07:02 AM