Skills
skills/knoopx/pi/nu-shell/Gen Agent Trust Hub

nu-shell

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documents how to execute Nushell scripts and inline commands using the nu CLI, such as nu -c 'ls | length'.
  • [DATA_EXFILTRATION]: The skill illustrates file interaction via the open and save commands, allowing the reading of structured data (JSON, CSV) and writing to the file system.
  • [PROMPT_INJECTION]: The skill defines a data processing environment that could be vulnerable to indirect prompt injection when handling external files.
  • Ingestion points: SKILL.md documents the use of the open command to ingest data from external files like config.json and data.csv.
  • Boundary markers: No markers or instructions for isolating untrusted data are provided in the documentation.
  • Capability inventory: SKILL.md documents file-writing capabilities (save) and shell command execution (nu -c).
  • Sanitization: The documentation does not describe methods for sanitizing or validating ingested file content.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 02:48 AM