ov-memory-doctor
Fail
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the user to execute a remote installation script via
curl -fsSL https://openviking.ai/install | bash. While this targets a domain associated with the skill's vendor, piping remote scripts to a shell is a sensitive operation.- [COMMAND_EXECUTION]: The skill primarily functions by executing a local Node.js script (node ${CLAUDE_PLUGIN_ROOT}/scripts/ov-memory-doctor.mjs) to perform diagnostics and provides various CLI commands for the user to manually verify server health and configuration.- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface because it ingests and processes external data, including server status responses and local execution logs, which could be influenced by external actors. - Ingestion points: Reads local diagnostic logs (
~/.openviking/logs/cc-hooks.log) and fetches data from external API endpoints (/health,/ready,/mcp). - Boundary markers: Relies on structured data formats (JSON) for status checks and configuration management, which helps mitigate accidental interpretation of data as instructions.
- Capability inventory: The skill can execute local scripts, perform network requests, and update configuration files based on the diagnostic results.
- Sanitization: The instructions include explicit rules for the agent to never print full API keys and to use masked forms provided by the diagnostic tool.
Recommendations
- HIGH: Downloads and executes remote code from: https://openviking.ai/install - DO NOT USE without thorough review
Audit Metadata