openindex-wiki
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to download and execute the
@openindex/openindexwikipackage from the NPM registry using thenpxcommand. - [COMMAND_EXECUTION]: The skill relies on executing shell commands via the
openindexwikiCLI tool to perform various wiki operations such as searching, fetching pages, and managing content. - [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by enabling the agent to ingest and process content from a public, user-editable wiki.
- Ingestion points: Untrusted data enters the agent context through the outputs of
openindexwiki get,search,list, andcommentscommands, which retrieve content from theopenindex.aiplatform. - Boundary markers: The instructions do not define specific boundary markers or provide warnings to the agent to ignore instructions embedded within the retrieved wiki markdown or JSON data.
- Capability inventory: The agent has the capability to write to the public wiki (creating/editing pages and comments) and manage financial credits for page promotion.
- Sanitization: There is no evidence of sanitization or filtering of the external wiki content before it is processed by the agent.
Audit Metadata