Cursor rule

1---
2description: "SAFe methodology core principles, round-table philosophy, search-first-reuse-always, evidence-based delivery. Apply to ALL work in this repository."
3alwaysApply: true
4---
5 
6# SAFe Core Principles
7 
8This repository uses **SAFe (Scaled Agile Framework) Agentic Workflow** with 11 specialized AI agent roles. You are part of a collaborative team where every contributor has equal voice.
9 
10## Round Table Philosophy
11 
12- Equal voice, mutual respect, shared responsibility
13- Every agent and human contributor's input carries equal weight
14- "Stop-the-line" authority: ANY contributor can halt work for architectural or security concerns
15- Decisions are evidence-based, not authority-based
16 
17## Search First, Reuse Always
18 
19**MANDATORY before implementing anything:**
20 
211. Search `patterns_library/` for existing code patterns
222. Search `specs/` for similar specifications
233. Search the codebase for similar implementations
244. Consult documentation (see CONTRIBUTING.md, docs/)
255. Only create new patterns if nothing exists (and that is BSA/Architect's job)
26 
27See `patterns_library/README.md` for the full pattern index (18+ patterns across 7 categories).
28 
29## SAFe Hierarchy
30 
31All work follows: **Epic > Feature > Story > Enabler**
32 
331. BSA creates spec in `specs/{{TICKET_PREFIX}}-XXX-feature-spec.md`
342. System Architect validates the architectural approach
353. Implementation agents execute using discovered patterns
364. QAS validates against acceptance criteria
375. Evidence is attached to the project management ticket before POPM review
38 
39## Metacognitive Tags
40 
41Use these in specs to flag critical decisions:
42 
43- `#PATH_DECISION` -- Architectural path chosen (document alternatives)
44- `#PLAN_UNCERTAINTY` -- Areas requiring validation
45- `#EXPORT_CRITICAL` -- Security or compliance requirements
46 
47## Key Resources
48 
49- `CLAUDE.md` -- Full AI assistant context and development commands
50- `AGENTS.md` -- All 11 agent roles, invocation patterns, exit states
51- `CONTRIBUTING.md` -- Git workflow, commit standards, PR process
52- `patterns_library/` -- Reusable code patterns (copy-paste ready)
53- `docs/onboarding/` -- Setup guides and daily workflows
54- `docs/guides/ROUND-TABLE-PHILOSOPHY.md` -- Collaboration principles
55 
56## Exit States
57 
58Every agent role has an explicit exit state defining its handoff point:
59 
60| Role             | Exit State                                |
61|------------------|-------------------------------------------|
62| BE-Developer     | "Ready for QAS"                           |
63| FE-Developer     | "Ready for QAS"                           |
64| Data-Engineer    | "Ready for QAS"                           |
65| QAS              | "Approved for RTE"                        |
66| RTE              | "Ready for HITL Review"                   |
67| System Architect | "Stage 1 Approved - Ready for ARCHitect"  |
68| HITL             | MERGED                                    |
69 
70## Quality Gates
71 
72| Gate            | Owner            | Blocking?                |
73|-----------------|------------------|--------------------------|
74| Stop-the-Line   | Implementer      | YES -- no AC = no work   |
75| QAS Gate        | QAS              | YES -- no approval = stop|
76| Stage 1 Review  | System Architect | YES -- pattern check     |
77| Stage 2 Review  | ARCHitect-CLI    | YES -- architecture check|
78| HITL Merge      | {{AUTHOR_NAME}}  | YES -- final authority   |
79 
80## Evidence-Based Delivery
81 
82- All work requires evidence attached to the project management ticket
83- Evidence includes: command output, test results, validation screenshots
84- No work is "done" without verifiable evidence
85