skills/glebis/claude-skills/tdd/Gen Agent Trust Hub

tdd

Pass

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • COMMAND_EXECUTION (LOW): The skill utilizes a test runner wrapper (scripts/run_tests.sh) to execute framework-specific test commands. As a TDD tool, this capability is intended; however, it provides a surface for executing any code generated by the sub-agents or specified in the test command.
  • PROMPT_INJECTION (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8) because it ingests external project data to inform its agents.
  • Ingestion points: The orchestrator reads README.md, API.md, and source code files using scripts/extract_api.sh and scripts/discover_docs.sh.
  • Boundary markers: Data is isolated using markdown headers (e.g., ## Project documentation) in the agent prompts, but there are no explicit instructions to ignore embedded commands within that data.
  • Capability inventory: The agents have the capability to write/overwrite source files and execute shell commands via the orchestrator.
  • Sanitization: There is no evidence of sanitization for docstrings or documentation content before they are included in the prompt context.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 23, 2026, 06:47 AM