meilisearch
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a technical guide for Meilisearch integration, focusing on configuration, performance tuning, and security best practices without introducing malicious code or settings.
- [CREDENTIALS_UNSAFE]: The provided code snippets and CLI recipes use a placeholder value 'MASTER_KEY_SECRET' for configuration. The skill correctly directs the agent to use short-lived, filter-scoped tenant search keys instead of master keys for frontends, adhering to security best practices.
- [EXTERNAL_DOWNLOADS]: The skill references official Meilisearch ecosystem resources, such as the 'meilisearch' NPM package and the 'getmeili/meilisearch' Docker Hub image, which are standard for the engine's implementation.
- [COMMAND_EXECUTION]: The skill includes Docker and cURL commands for standard operational tasks like starting the engine, checking task status, and creating database dumps.
- [INDIRECT_PROMPT_INJECTION]: The skill provides example code that searches an index and prints results, representing a potential injection surface if the searched data is malicious. 1. Ingestion points: Meilisearch search results ('searchResults.hits') in SKILL.md. 2. Boundary markers: Absent in the example log output. 3. Capability inventory: Use of 'console.log' in SKILL.md. 4. Sanitization: Absent; hit attributes are directly interpolated into strings. This finding is considered safe as it appears in a non-executable educational context.
Audit Metadata