skills/algolia/skills/instantsearch/Gen Agent Trust Hub

instantsearch

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes curl for index schema discovery by querying the Algolia API. This is a standard and necessary operation for identifying available data attributes, performed using credentials provided by the user to access the vendor's official infrastructure.
  • [EXTERNAL_DOWNLOADS]: The skill directs the agent to install search libraries and fetch live documentation from verified vendor-owned domains and standard package registries. These resources are trusted and essential for implementing the search functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a process for ingesting search records to propose rendering options and UI components based on the data schema.
  • Ingestion points: Algolia index hits are fetched via curl as specified in the Discover workflow in SKILL.md.
  • Boundary markers: The agent processes the JSON structure of the records; specific prompt boundary markers are not utilized for this discovery task.
  • Capability inventory: The skill includes operations for file writing (code scaffolding), dependency installation, and network requests via curl.
  • Sanitization: The agent performs structural analysis of the JSON response to identify record attributes.
  • Note: This ingestion is a primary and documented function of the skill, targeting user-controlled data to facilitate development, and does not pose an atypical security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:41 AM
Security Audit — agent-trust-hub — instantsearch