project-context-ingestion

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill identifies and analyzes standard project configuration files such as Gradle build scripts, Spring properties, and Docker manifests. This is a core function used to determine technical compatibility and does not involve unauthorized data access or transmission.
  • [PROMPT_INJECTION]: The skill ingests untrusted data from the repository's source files, which presents a surface for indirect prompt injection. Ingestion points: Build files, properties, and manifests in the target repository. Boundary markers: Not explicitly implemented. Capability inventory: No file-write, shell execution, or network communication capabilities across the provided scripts. Sanitization: No specific filtering of ingested text is performed.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 07:32 PM
Security Audit — agent-trust-hub — project-context-ingestion