testing-setup

Fail

Audited by Gen Agent Trust Hub on Oct 5, 2026

Risk Level: HIGHMETADATA_POISONINGINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [METADATA_POISONING]: The skill's frontmatter provides misleading information by claiming "Google LLC" as the author, which is inconsistent with the provided author identity of "deonmenezes". Additionally, the metadata includes a "last-updated" date in the future (2026-09-23), further indicating unreliable metadata.
  • [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting and processing data from the project environment which could contain malicious instructions.
  • Ingestion points: The skill instructions direct the agent to analyze local project files including libs.versions.toml, Gradle build scripts, and application source code (ViewModels, Repositories).
  • Boundary markers: No specific delimiters or safety instructions are defined to separate untrusted project content from the agent's core instructions, increasing the risk of the agent obeying instructions embedded in the analyzed code.
  • Capability inventory: The agent has the capability to write to the file system, including creating new unit tests, modifying build configurations to install libraries, and generating or updating documentation files like AGENTS.md.
  • Sanitization: The skill does not implement any validation or sanitization steps for the data retrieved from the local project environment before it is used in automated tasks.
  • [EXTERNAL_DOWNLOADS]: The skill references external resources from a well-known technology service to facilitate the setup of testing tools.
  • Evidence: Accesses documentation content from the official repository of a well-known service at https://raw.githubusercontent.com/dropbox/dropshots/refs/heads/main/README.md.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Oct 5, 2026, 09:27 PM