ai-dev-team

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists of markdown instructions defining a development process contract. It does not include executable scripts, hardcoded credentials, or unauthorized network operations.
  • [PROMPT_INJECTION]: The workflow design involves agents reading task descriptions and project documentation from external markdown files (e.g., Backlog.md). This creates a surface for indirect prompt injection where instructions embedded in these files could influence agent behavior.
  • Ingestion points: Backlog.md and project-specific knowledge base files referenced in the workflow.
  • Boundary markers: The contract does not define explicit delimiters or instructions for the agent to ignore embedded commands within these external files.
  • Capability inventory: Agents following this contract are expected to perform file edits, execute tests via verify-landed, and manage git operations (branching, committing, and opening PRs).
  • Sanitization: The skill does not describe any specific sanitization or validation steps for the content ingested from the backlog or knowledge base.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 07:35 PM
Security Audit — agent-trust-hub — ai-dev-team