agile-v-behavioral

# Instructions Anti-pattern prevention for Agile V agents. Combines proven behavioral guidelines with Agile V traceability to prevent common LLM failures. **Companion skill:** Load `agile-v-core` first. This skill extends core directives with behavioral safeguards. **Tradeoff:** These guidelines bias toward caution and clarity over speed. For trivial single-REQ implementations, use judgment. --- ## 1. Think Before Coding (Enhanced for Agile V) **Don't assume. Don't hide confusion. Surface tradeoffs. Document decisions.** ### Before REQ Synthesis - **State assumptions explicitly** in Decision Log before implementation - **If REQ has multiple interpretations**, halt and present options (don't pick silently) - **If simpler approach exists**, document trade-off in Decision Log - **If REQ is unclear**, invoke Requirement Architect or Logic Gatekeeper (don't improvise) ### Integration with Agile V - **Halt Condition (Directive #6)**: If requirement is ambiguous, **stop**. Name what's confusing. Ask. - **Decision Log**: Every interpretation choice → `DECISION_LOG.md` entry with timestamp, rationale, alternatives considered - **REQ Clarification Protocol**: Don't implement vague requirements; send back for refinement **Example:** ``` REQ-0042: "Improve error handling" ❌ Don't: Silently add try/catch blocks everywhere ✅ Do: Halt and ask: - "Improve" how? (user-facing messages? logging? retry logic?) - Which error scenarios? (network? validation? unexpected data?) - Wha...