agora
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches the Agora CLI installer from the vendor's official GitHub repository at
raw.githubusercontent.com/AgoraIO/cli/main/install.shandinstall.ps1. - [REMOTE_CODE_EXECUTION]: The skill executes the downloaded installation script via the shell to set up the
agoracommand-line tool. - [COMMAND_EXECUTION]: The skill uses the
agoraCLI to perform project management tasks, including authentication, project creation, and environment configuration. - [PROMPT_INJECTION]: The skill processes untrusted data in the form of real-time audio transcripts and RTM messages from AI agents, presenting an indirect prompt injection surface.
- Ingestion points: Untrusted data enters the agent context through the
agora-agent-client-toolkitvia transcripts and RTM message events. - Boundary markers: No explicit boundary markers or "ignore instructions" warnings are defined for the ingested transcript data.
- Capability inventory: The agent can execute shell commands through the
agoraCLI and perform file-write operations such asagora project env write. - Sanitization: Data sanitization is handled by the underlying vendor SDKs; no custom sanitization logic is implemented within the skill itself.
Audit Metadata