iblai-vibe-history
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes and renders conversation transcripts (AI and user messages), which constitutes an ingestion point for untrusted data.
- Ingestion points: Conversation data is retrieved via
useGetUserChatHistoryQueryanduseGetUserChatHistorySessionQuery(referenced in SKILL.md). - Boundary markers: The UI renders these messages as Markdown in the transcript preview column.
- Capability inventory: The skill is a UI component; no backend execution, file system writes, or network exfiltration capabilities are triggered by the content of these messages in the documented implementation.
- Sanitization: Markdown formatting is automatically stripped when rendering list rows to prevent injection into the list view navigation.
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download an environment configuration template from a remote repository.
- Evidence:
curl -o iblai.env https://raw.githubusercontent.com/iblai/vibe/refs/heads/main/iblai.env(SKILL.md) - Analysis: This operation fetches a non-executable configuration template from the vendor's official repository for environment setup purposes.
Audit Metadata