algolia-agent-studio
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill consists of documentation and configuration guidelines for Algolia Agent Studio. No executable code or scripts are included in the repository.
- [EXTERNAL_DOWNLOADS]: The skill references implementation guidance from official Algolia documentation and Academy domains (algolia.com, academy.algolia.com). These are trusted vendor resources.
- [INDIRECT_PROMPT_INJECTION]: The skill identifies a vulnerability surface where agents ingest search hits and turn context. Ingestion points: Search hits and turn context (references/agent-studio-guide.md). Boundary markers: Instructions advise on 'custom guardrails' and 'refusal policy'. Capability inventory: Search tool and Recommend tool access. Sanitization: Instructions recommend 'validate arguments' and 'least-privilege credentials'.
- [CREDENTIALS_UNSAFE]: The instructions explicitly forbid the use of admin API keys or unrestricted credentials in client-facing agents.
Audit Metadata