searching-context
Pass
Audited by Gen Agent Trust Hub on Feb 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various commands using the
ovCLI utility, includingov search,ov grep,ov glob,ov ls,ov tree, andov read. These commands are used to interact with the local OpenViking context and configuration located at~/.openviking/ovcli.conf. - [PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection as it retrieves data from external resources and memories.
- Ingestion points: Data enters the agent's context through search results and resource content retrieved via
ov search,ov grep, andov read. - Boundary markers: The documentation does not define specific delimiters or instructions to treat the retrieved content as untrusted data.
- Capability inventory: The skill provides full content retrieval (
ov read), directory traversal (ov ls,ov tree), and recursive search capabilities via theovCLI. - Sanitization: There is no indication that the skill sanitizes or validates the data retrieved from OpenViking resources before passing it to the agent.
Audit Metadata