skills/plurigrid/asi/abductive-repl/Gen Agent Trust Hub

abductive-repl

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION] (MEDIUM): The skill utilizes julia -e within its Justfile to execute arbitrary strings as code. It also provides a REPL interface (Gay.repl()) which is inherently designed for dynamic execution. This poses a risk if an attacker can influence the parameters passed to these commands.
  • [EXTERNAL_DOWNLOADS] (MEDIUM): The skill depends on a Julia project (Gay.jl) and a Python package (abductive_repl) that are not from the provided list of trusted sources. These dependencies are unverifiable.
  • [DATA_EXPOSURE] (LOW): The configuration specifies a history file located at ~/.abductive_history. While standard for REPLs, it represents file system interaction with the user's home directory.
  • [INDIRECT_PROMPT_INJECTION] (MEDIUM): The skill processes external input (RGB values, IDs) through its commands and recipes. There is a lack of explicit sanitization or boundary markers shown in the provided documentation, creating a potential surface for injection if these inputs are interpolated into shell commands or REPL evaluations.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 11:18 AM