spring-data-neo4j

Pass

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: SAFECREDENTIALS_UNSAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE]: Configuration examples in SKILL.md and references/reference.md include dummy hardcoded credentials (e.g., password=secret). While intended as placeholders for documentation, they follow patterns for insecure credential storage.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface through data retrieved from Neo4j databases.
  • Ingestion points: Data retrieved via MovieRepository and UserRepository in references/examples.md.
  • Boundary markers: Not specified in the integration patterns.
  • Capability inventory: Full CRUD operations on graph databases and command line access via Bash.
  • Sanitization: The skill demonstrates best practices by using parameterized queries ($param) to prevent Cypher injection into the database, but does not sanitize retrieved content for potential LLM instructions.
  • [COMMAND_EXECUTION]: The skill requires the Bash tool to execute standard Maven and Gradle build commands and environment setup tasks.
  • [EXTERNAL_DOWNLOADS]: Fetches development dependencies and configuration from official Spring and Neo4j repositories via Maven and Gradle.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 23, 2026, 11:17 PM