skills/garrytan/gstack/design-review/Gen Agent Trust Hub

design-review

Fail

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches the Bun installer from its official domain (bun.sh) if the runtime is missing during setup.
  • [REMOTE_CODE_EXECUTION]: Executes the Bun installation script after verifying its integrity with a hardcoded SHA256 checksum.
  • [COMMAND_EXECUTION]: Utilizes bash commands and local binaries (e.g., gstack-config, gstack-slug) to manage project sessions, configuration, and telemetry within the ~/.gstack and ~/.claude directories.
  • [PROMPT_INJECTION]: The skill ingests untrusted content from the web via search and browsing tools, creating a surface for indirect prompt injection.
  • Ingestion points: External websites retrieved via the browse tool and WebSearch.
  • Boundary markers: None identified in the instruction set.
  • Capability inventory: File system modification, shell command execution, and git operations.
  • Sanitization: No explicit sanitization of external content is performed.
Recommendations
  • HIGH: Downloads and executes remote code from: https://bun.sh/install - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
May 20, 2026, 02:59 PM
Security Audit — agent-trust-hub — design-review