skills/appwrite/skills/appwrite-php/Gen Agent Trust Hub

appwrite-php

Pass

Audited by Gen Agent Trust Hub on Oct 23, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill recommends installing the appwrite/appwrite package via Composer. This is the official library provided by the vendor.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface where untrusted data could influence agent behavior during database or function operations.
  • Ingestion points: Untrusted data enters the context through createRow data arrays, createFile paths, and the $context->req properties (like bodyJson) in the serverless function handler script (src/main.php).
  • Boundary markers: The code examples do not include explicit delimiters or warnings to treat processed data as non-instructional.
  • Capability inventory: The skill utilizes the Appwrite SDK to perform file system operations (InputFile), persistent database writes (TablesDB), and serverless logic execution (Functions).
  • Sanitization: No input validation or sanitization logic is demonstrated in the simplified code snippets, leaving the risk of processing malicious payloads if user-supplied data is not handled carefully by the implementation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 23, 2026, 05:53 AM