knowledge-patch-setup

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Python helper script knowledge_patch_setup.py to perform project scanning and management of skills and hooks.
  • [EXTERNAL_DOWNLOADS]: The skill downloads knowledge patch skills from the official Nevaberry/nevaberry-plugins GitHub repository. This behavior is the primary intended function of the skill and originates from the vendor's own infrastructure.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection (Category 8) because it processes untrusted data from the local project root.
  • Ingestion points: The project root is scanned by knowledge_patch_setup.py to detect technologies based on file content and structure.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore embedded instructions within the scanned project files.
  • Capability inventory: The skill can execute shell commands, install new agent skills into $CODEX_HOME/skills, and modify hook configuration files (hooks.json).
  • Sanitization: The skill does not describe any sanitization or validation of the content discovered during the project scan before using it to decide which patches to install.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 02:01 PM