quality-stack

Pass

Audited by Gen Agent Trust Hub on Apr 26, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches research documentation in Markdown format from the author's public GitHub repository (github.com/joaquimscosta/arkhe-claude-plugins). These documents are used to cross-reference detected tools with industry recommendations and provide setup instructions.
  • [COMMAND_EXECUTION]: The skill executes local Python scripts included in the scripts/ directory to perform project scanning. Additionally, it provides the agent with instructions to execute standard development commands (e.g., ./gradlew, npm, pnpm, uv) to install and configure tools based on user selection.
  • [PROMPT_INJECTION]: The skill processes external data from local project files (e.g., build.gradle, package.json, pyproject.toml) and remote research documents to determine its actions.
  • Ingestion points: Local project configuration files and remote Markdown documents fetched via WebFetch.
  • Boundary markers: None explicitly defined for separating processed project data from instruction context.
  • Capability inventory: File system modification and shell command execution for tool setup.
  • Sanitization: Employs manual 'Setup Guards' and post-setup verification commands to ensure generated configurations are valid and functional.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 26, 2026, 10:55 AM
Security Audit — agent-trust-hub — quality-stack