skills/agentivestack/skills/tdd/Gen Agent Trust Hub

tdd

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and follow project-specific documentation, which could be exploited if an attacker provides a malicious codebase.
  • Ingestion points: Documentation files (docs/contexts/<name>/CONTEXT.md, docs/UBIQUITOUS_LANGUAGE.md) and user-provided arguments (issue numbers, slice descriptions) are used to define system behavior.
  • Boundary markers: There are no explicit markers or instructions for the agent to ignore or sanitize embedded instructions within these external files.
  • Capability inventory: The agent has the capability to write files and execute test runners (shell commands), creating a path for code execution based on the ingested documentation.
  • Sanitization: No validation or sanitization is performed on the content read from the project documentation before it influences the agent's actions.
  • [DYNAMIC_EXECUTION]: The skill's primary function is the creation and execution of scripts (test files and implementation code) at runtime.
  • Evidence: The "Tracer bullet" and "Incremental loop" processes explicitly require the agent to write new code and execute it using the project's test infrastructure to verify success or failure.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 02:42 PM
Security Audit — agent-trust-hub — tdd