just
Pass
Audited by Gen Agent Trust Hub on May 21, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: Provides instructions to install the tool by piping a script from the official 'just.systems' domain directly into the shell. This is the vendor's recommended installation method.\n- [EXTERNAL_DOWNLOADS]: Fetches installation resources and scripts from the tool's official website during the setup process.\n- [PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection because the agent is instructed to read and execute commands from a 'justfile', which could contain malicious instructions embedded in recipe descriptions or shell snippets.\n
- Ingestion points: Project-level
justfilecontents read viajust --listorjust --show.\n - Boundary markers: No instructions are provided to the agent to treat file content as untrusted data or to ignore instructions embedded within the file.\n
- Capability inventory: The skill allows the agent to execute shell recipes, format files, and view variables using the
justCLI.\n - Sanitization: No sanitization or validation logic is defined for the content of the recipes before execution.
Audit Metadata