iblai-vibe-agent-history
Warn
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSNO_CODE
Full Analysis
- [COMMAND_EXECUTION]: Verification instructions include a shell command that interpolates a user-provided
mentorIddirectly into a URL for Playwright, which could lead to command injection if the identifier contains shell metacharacters. - Evidence:
npx playwright screenshot http://localhost:3000/agents/<id>/history /tmp/agent-history.pngwhere<id>is a value the agent is instructed to request from the user (SKILL.md). - [INDIRECT_PROMPT_INJECTION]: The
AgentHistoryTabcomponent renders untrusted conversation history, creating a potential surface for indirect instructions to influence the agent. - Ingestion points: The component processes and renders message content via the
renderMessageContentprop (SKILL.md). - Boundary markers: The implementation examples do not provide delimiters or specific instructions for the agent to ignore instructions embedded within the conversation data.
- Capability inventory: The skill instructions grant the agent capabilities to execute build processes (
pnpm build), development servers (pnpm dev), and browser automation (npx playwright) (SKILL.md). - Sanitization: Uses
ReactMarkdownfor content rendering, which provides structural sanitization but does not filter out text-based instructions targeted at an agent reading the rendered output. - [EXTERNAL_DOWNLOADS]: Fetches environment configuration templates and brand guidelines from the vendor's official GitHub repository.
- [NODE_PACKAGES]: Depends on vendor-maintained packages (
@iblai/iblai-js,@iblai/iblai-web-mentor,@iblai/mcp) and common utility libraries (sonner,react-markdown,playwright).
Audit Metadata