agora
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill provides instructions for installing the Agora CLI using remote shell scripts (
curl -fsSL https://dl.agora.io/cli/install.sh | shandirm https://dl.agora.io/cli/install.ps1 | iex). These resources are hosted on the official vendor domaindl.agora.ioand are used for administrative setup. - [EXTERNAL_DOWNLOADS]: The skill references numerous external resources, including official Agora SDKs on npm, PyPI, and Maven, and sample repositories on GitHub (
github.com/AgoraIO,github.com/AgoraIO-Conversational-AI). These are verified vendor resources. - [COMMAND_EXECUTION]: The skill utilizes the Agora CLI (
agorabinary) to perform tasks such as logging in, creating projects, enabling features, and diagnosing setup issues viaagora doctorandagora project doctor. It also mentions spawning child processes in the custom LLM server implementation. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it fetches documentation from external sources (
docs.agora.io,llms.txt) and processes user-provided project configuration files. The risk is considered low due to the structured workflow and reliance on official reference files.
Audit Metadata