chat-customizations-editor

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use npx @playwright/cli eval to execute JavaScript snippets within the browser context for debugging layout and inspecting widgets. These actions are consistent with the skill's purpose as a development tool for the management UI.
  • [PROMPT_INJECTION]: The skill facilitates the management of potentially untrusted data such as AI skills, prompts, and hooks, creating a surface for indirect prompt injection.
  • Ingestion points: The UI processes IAgentSkill objects, IChatPromptSlashCommand definitions, and prompt files retrieved via listPromptFiles as described in SKILL.md.
  • Boundary markers: The provided documentation does not specify the use of delimiters or boundary markers to separate customization content from system instructions.
  • Capability inventory: The skill allows for command execution via Bash for running tests (./scripts/test.sh) and UI automation tools (npx @playwright/cli).
  • Sanitization: The documentation focuses on UI components and does not describe validation or sanitization routines for the processed customization data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 01:46 PM
Security Audit — agent-trust-hub — chat-customizations-editor