v3.5.0 — E2E Quality Gate + Mini-App Chấm Công Showcase

Release v3.5.0 — E2E Quality Gate + Mini-App Chấm Công Showcase + Anti-Rationalization Unified

Date: 2026-04-14
Marketing version: v3.5.0
CHANGELOG version: 1.5.0
Branch: feat/mini-app-cham-cong-docs → main (merged via --no-ff)
Merge commit: c9c31e3 (local, not pushed)
Tag: v3.5.0 (local, not pushed)
Agent count post-merge: 44 (33 with anti-rationalization + 11 sprint spine)

Reconciliation note: This release merges TWO parallel development tracks
that diverged from cabccd6 (2026-04-12). The initial intent was a simple
merge of the feat branch into main, but surveying revealed main had shipped
significant work (v3.2 + v3.3 Anti-Rationalization Framework, license
change, docs rewrite) that feat branch never absorbed. Both tracks are now
unified into v3.5.0.

Reconciliation audit

Stat	Value
Feat branch commits (absorbed)	15 (14 original + 1 release prep)
Main branch commits (absorbed)	9
Files in overlap (both sides modified)	104
Hand-resolved UU conflicts	11 (4 root + 6 docs/ + 1 ba-master SKILL)
Auto-merged SKILL.md files	25
Modify-delete conflicts resolved (keep feat)	65 (outputs/mini-app-cham-cong/*)
Smoke test verdict	PASS — 243/0/0/4/247

---

TL;DR

Two major capabilities land on top of the v3.4 Sprint Spine:

1. Phase 08 CI quality gate — BA artifacts (skills, docs, fixtures) now validated on every PR via GitHub Actions, 5-layer lint orchestrator, JSON + Markdown reports
2. Mini-App Chấm Công end-to-end showcase — 12 modules, 4 BRDs, 53 US files, EAMS v2.1, RTM, 111/111 Confluence pages — proves the kit dogfooded on a real enterprise attendance system

Zero breaking changes. Drop-in upgrade from v3.4.0.

---

Release highlights

1. CI-integrated E2E quality gate (Phase 08)

Orchestrator — .agent/scripts/ba_e2e_test.py (1,389 LOC, stdlib only)

• 5 lint layers:
  • L1 — script smoke (syntax/import for all ba_*.py helpers)
  • L2 — helper stdlib interface smoke (14 test cases covering ba_as_built, ba_retro, ba_learn, ba_baseline, ba_second_opinion, ba_setup)
  • L3 — skill frontmatter + canonical sections (Input Validation, System Instructions, etc.)
  • L4 — deep BRD↔US↔API↔test consistency (6 lint families: duplicate US IDs, BRD→US xref, API-spec US refs, test-case US refs, RTM US refs, orphaned test cases)
  • L5 — fixture coverage (US ID extraction, lowercase tolerance)
• --report markdown output, --json stable schema v1.0.0 (ensure_ascii=False for Vietnamese round-trip), --fixture selector

Workflow — .github/workflows/e2e-skills.yml

• Triggers on PR + push to main/feat/** touching .agent/, docs/, outputs/, or the workflow itself
• Python 3.11, actions/checkout@v4 fetch-depth: 0 (ba_retro needs git history)
• Writes markdown to GITHUB_STEP_SUMMARY for inline PR review UX
• Uploads reports/ artifact (14-day retention)
• Sticky PR comment via marocchino/sticky-pull-request-comment@v2
• Fails job only on FAIL / CRASH verdict — WARN doesn't block
• 10-min timeout, zero external deps

README badge — GitHub Actions status badge linked to e2e-skills.yml, now tracking main branch.

2. Mini-App Chấm Công — comprehensive BA showcase

Location: outputs/mini-app-cham-cong/

Overview artifacts

• 4 BRDs — Nhân viên / Quản lý / HR-Admin / IT-System-Admin
• Demo Plan Sprint 8
• Modules overview
• Confluence manifest (111 pages)

12 modules (M01-M12) — each with README.md + api-spec.md + db-schema.md + test-cases.md + US files:

• M01 Chấm công — check-in/out, GPS, selfie, anti-spoofing
• M02 OT — overtime request/approval/settlement
• M03 Nghỉ phép — leave types, balance, approval chain
• M04 Công tác — business trip, expense
• M05 Phân ca — shift scheduling, rotation, swap
• M06 Chính sách — policy engine, rules
• M07 Mobile — offline sync, push, deep-link
• M08 (reserved)
• M09 (reserved)
• M10 (reserved)
• M11 Báo cáo — weekly, monthly, compliance, payroll lock
• M12 Quản trị hệ thống — chi nhánh, audit log, offboarding, chốt công tháng, onboarding, data retention

User stories — 53 total

• M01-M11: 47 US files rewritten R3 with concrete Gherkin scenarios sourced from Edge Cases tables
• M12: 6 US files (us-sys-01 → us-sys-06)

Supporting docs

• eams-v2-comprehensive.md — EAMS v2.1 (§15.3 attendance anomalies, §17.5 leave accrual, §17.6 payroll lock)
• RTM.md — full requirements traceability matrix
• AUDIT-REPORT.md — R3 remediation audit trail
• INDEX.md — navigation entry

Confluence sync — 111/111 pages pushed, 0 rendering errors. Unblocked by language-mapping fix (json/gherkin) + mermaid-macro support in push scripts.

3. Skill hygiene

• ba-wiki/SKILL.md — canonical ## Input Validation + ## System Instructions added. L3 frontmatter lint 6/8 → 8/8 (operation-based template was missing headers because it uses Ingest/Query/Lint instead of the cognitive-loop template)
• coverage_checker.py — accepts lowercase us-* + test-cases.md filenames

---

Commit log (14 commits)

SHA	Summary
9c79110	feat(ci): Phase 08 — JSON reporter + GitHub Actions workflow + badges
00032e8	fix(skills): ba-wiki add canonical Input Validation + System Instructions
a8bb262	feat(scripts): ba_e2e_test L4 deep lints — BRD↔US↔API↔test consistency
7c7be35	feat(scripts): ba_e2e_test L2 — script helper smoke tests (14 tests)
4cd41e1	feat(scripts): ba_e2e_test.py orchestrator — L1 + L3 + L5 layers
682e8bd	fix(scripts): coverage_checker recognizes lowercase us-* and test-cases.md
6784853	chore: rename release v3.2.0 → v3.4.0 (tag conflict reconciliation)
2fb7392	docs: v3.2 release refresh — badges, sprint spine, cheat-sheet, changelog
9d4821f	feat: v3.2 sprint spine — 11 new agents + 6 helpers + setup wizard
e1a0625	fix: YAML frontmatter quotes + BA-fit polish across 26 skills
0420b82	docs: Confluence sync complete — 111/111 pages updated, 0 rendering errors
2823845	fix: Confluence DC rendering blockers — json/gherkin lang mapping + mermaid-macro
37cc552	docs: R3 Gherkin remediation — M01-M11 (47 US files)
3eb75b8	docs: R3 remediation — M12 Gherkin rewrite (6 US), EAMS v2.1, API/BRD/RTM fixes

Plus the release prep commit (added on merge): chore(release): v3.5.0 — CHANGELOG + badges

---

Upgrade path

From v3.4.0 — drop-in, no breaking changes. To activate the new CI quality gate on a fork:

1. git pull origin main
2. Ensure .agent/scripts/ba_e2e_test.py is present
3. Copy .github/workflows/e2e-skills.yml into your repo (or fork will inherit on first PR)
4. First PR touching .agent/, docs/, or outputs/ triggers the gate automatically

From <v3.4.0 — see CHANGELOG.md §1.4.0 for Sprint Spine migration (11 new agents, ba-setup wizard, BA-fit reframing).

---

Post-merge checklist (user action)

Current local state (as of 2026-04-14 03:15):

• main is at c9c31e3 (merge commit) — 14 commits ahead of origin/main
• v3.5.0 tag exists locally, not pushed
• feat/mini-app-cham-cong-docs branch still exists (same commit as main now)

Review before pushing:

• git log main --oneline -20 — review the reconciled history
• git show v3.5.0 --stat — verify tag points to merge commit
• git diff origin/main main --stat | tail -5 — sanity-check LOC delta
• python3 .agent/scripts/ba_e2e_test.py --fixture outputs/mini-app-cham-cong — re-run smoke test
• Open README.md visually — check badge count says 44, logo cache-bust, sprint spine section intact
• Open CHANGELOG.md — verify §1.5.0 reconciliation note + historical main entries preserved
• Spot-check 2-3 SKILL.md files (e.g. ba-elicitation, ba-writing, ba-validation) — verify anti-rationalization sections AND sprint-spine context both present

Push sequence (when ready):

git push origin main                    # trigger CI on main, badge should flip green
git push origin v3.5.0                  # publish tag
gh release create v3.5.0 \
  --title "v3.5.0 — E2E Quality Gate + Mini-App Chấm Công Showcase" \
  --notes-file plans/reports/release-260414-0237-v3-5-0.md

Optional cleanup:

• git push origin --delete feat/mini-app-cham-cong-docs — remote branch no longer needed
• git branch -d feat/mini-app-cham-cong-docs — local branch cleanup

Reconciliation trace

Why the scope ballooned — the initial "merge feat into main" plan assumed main was stale. Surveying revealed main had independently shipped:

• v3.2.0 Anti-Rationalization Framework (commit 766afb9) — 5 flagship agents + using-ba-kit meta skill + 6 slash commands
• v3.3.0 full rollout (commit b2444c1) — 28 more agents with When to Use / Common Rationalizations / Red Flags / Verification sections (~140 items each)
• License MIT → CC BY-NC 4.0 (commit 985d2d8)
• Logo asset refresh (commit e364cf5)
• outputs/ removed from tracking (commit b344068)
• knowledge-search domain mapping fixes (commit 125af14)

Critical conflict zones:

1. outputs/ modify-delete (65 files) — main had git rm -r outputs/, feat kept modifying outputs/mini-app-cham-cong/*. Resolved by: keeping feat versions + updating .gitignore from outputs/ to outputs/* + !outputs/mini-app-cham-cong/** negation.
2. 26 SKILL.md files — git auto-merged 25 successfully (different sections edited). Only ba-master/SKILL.md required hand-merge due to both sides editing the Decision Matrix routing table.
3. Root docs (README, CHANGELOG, AGENTS) — heavy manual reconciliation. CHANGELOG union both histories (feat's [1.x.y] semver + main's v3.x.y plain).
4. CHANGELOG.md — kept feat's 1.5.0 + 1.4.0 at top with a "Reconciliation note" block, appended main's v3.3.0/v3.2.0/v3.1.1/v3.1.0 entries below a "Historical entries" separa...

BA-Kit v3.3.0 — Full Anti-Rationalization Coverage

BA-Kit v3.3.0 — Full Anti-Rationalization Coverage

Pattern coverage: 5/33 → 33/33 agents (100%)

v3.2.0 shipped the anti-rationalization pattern on 5 flagship agents as exemplars. v3.3.0 completes the rollout — all 33 BA-Kit agents now include anti-slack enforcement mechanisms.

The Problem Being Solved

AI agents default to the shortest path. They rationalize skipping important steps:

• "The process is obvious" → skip BPMN diagram
• "Dev team will figure out performance" → skip NFR definition
• "Everything is a Must Have" → fake prioritization
• "NPV is too academic" → ROI hand-waving
• "A/B testing is for big tech" → skip hypothesis validation
• "Dashboards are for show" → surface-level audit

Each skipped step compounds. The result: incomplete specs, late-stage surprises, production incidents, stakeholder frustration.

The Solution

Every BA agent now enforces quality gates via 4 new sections:

Section	Purpose
When to Use / NOT to Use	Explicit triggers + exclusions. Prevents wrong-agent invocation.
Common Rationalizations	Table of excuses + factual rebuttals. Prevents skill-skipping.
Red Flags	Observable signs the skill is being violated. Enables self-monitoring.
Verification	Evidence-based exit criteria with handoff. Non-negotiable proof.

Scale of Change

Metric	Value
Agents updated this release	28
Total agents with anti-rationalization	33/33 (100%)
New domain-specific rationalizations	140+
New observable red flags	140+
New evidence-based verification items	140+
Lines added	1,170
Files modified	28

28 Agents Updated (4 Batches)

Batch A — Core Workflow Specialists:
ba-identity ba-nfr ba-process ba-prioritization ba-solution ba-conflict ba-export

Batch B — Advanced Analysis:
ba-metrics ba-root-cause ba-innovation ba-strategy ba-facilitation ba-systems ba-agile

Batch C — Quality & Modeling:
ba-test-gen ba-quality-gate ba-consistency ba-auditor ba-traceability ba-data ba-business-rules

Batch D — UX, Delivery & Integration:
ba-ux ba-communication ba-change ba-diagram ba-jira ba-confluence ba-wiki

Sample Rationalizations (Domain-Specific)

@ba-metrics (SPC/Cpk domain):

"Trend is up, we're fine" → "Up 3 points means nothing without UCL/LCL. Plot it on a control chart before declaring victory."

@ba-wiki (Knowledge management):

"Duplicate pages don't hurt anything" → "Duplicate pages create two sources of truth. When they diverge — and they always do — contradictions follow."

@ba-jira (Integration):

"Ticket title can differ from the US title" → "Title breaks traceability from the User Story. Keep them identical — or the RTM link breaks and @ba-traceability cannot trace it."

@ba-conflict (Harvard Negotiation):

"Compromise is the fair outcome" → "Compromise splits the difference. Innovation satisfies both parties' interests without splitting anything. Try Innovation first."

@ba-solution (Financial modeling):

"NPV is too academic, ROI is enough" → "Your CFO disagrees. NPV beats ROI when cash flows span >1 year."

Quality Assurance

• Automated check: 28/28 files pass section count verification (4 required sections each)
• Specificity test: Each rationalization is domain-specific (cannot be copy-pasted between agents)
• Observable test: Each red flag is checkable from the agent's output
• Evidence test: Each verification item requires commands, metrics, or file-level checks
• Handoff test: Last verification item references next agent in workflow

Methodology

Delivered via 4 parallel thematic subagents (7 files each) with pre-built domain seeds to prevent generic filler. Implementation plan: plans/260413-0832-anti-rationalization-rollout/plan.md

What's Next

With all 33 agents covered, the pattern is now the standard. Future agents created via the format in docs/skill-anatomy.md will include these sections by default.

Breaking Changes

None. All changes are additive.

Upgrade Path

No migration needed. Existing agent invocations continue to work. New sections are enforcement mechanisms — they help agents stay disciplined, not change behavior.

BA-Kit v3.2.0 — Anti-Rationalization Framework

BA-Kit v3.2.0 — Anti-Rationalization Framework

Production-grade engineering patterns for AI agents. BA-Kit now enforces the same discipline that prevents AI coding agents from rationalizing their way out of important steps.

The Problem

BA agents, like all AI agents, default to the shortest path. They rationalize skipping validation, skipping acceptance criteria, skipping test generation. Classic excuses:

• "The requirements are obvious"
• "Stakeholder already approved"
• "I'll add acceptance criteria later"
• "This story is too simple for INVEST"

The Solution — 4 New Anti-Slack Sections

Every flagship BA agent now includes:

1. When to Use / When NOT to Use — Explicit triggers and exclusions
2. Common Rationalizations — Table of excuses + factual rebuttals
3. Red Flags — Observable signs the skill is being violated
4. Verification — Exit criteria checklist with evidence requirements

Example: @ba-writing Rationalizations

Rationalization	Reality
"Acceptance criteria are obvious from the story"	Obvious to you. Dev will ship without them. Write 3 minimum: Happy + Edge + Error.
"I'll skip Gherkin, prose is fine"	Prose ACs fail automation. Gherkin forces precision.
"This story is too small for INVEST"	Smaller stories need MORE INVEST rigor, not less.
"Skip RBAC matrix, it's just one role"	Today's one role is tomorrow's four. Document now or regret later.

---

New Features

1. Lifecycle Slash Commands (Claude Code)

6 commands map directly to BA lifecycle phases:

Command	Phase	Agent Chain
/ba-discover	Discover	identity → strategy → questioning → elicitation → ux
/ba-analyze	Analyze	process / data / systems / business-rules
/ba-specify	Specify	writing → nfr → traceability → diagram
/ba-validate	Validate	validation → quality-gate → consistency → test-gen → auditor
/ba-deliver	Deliver	export → jira → confluence → communication → change
/ba-audit	Audit	auditor → traceability → consistency → metrics

Run the entire phase with one command. Commands live in .claude/commands/ba-*.md.

2. Meta Skill: using-ba-kit

A new meta skill for intent-to-skill routing. When an AI agent or user is unsure which BA skill applies, this skill:

• Parses user intent (phase + artifact + audience)
• Maps to the right agent via decision matrix (33 entries)
• Recommends a chain (not just single agent)
• Activates lifecycle slash commands
• Anti-slack enforcement

Now the fallback route for @ba-master when intent is unclear.

3. Skill Anatomy Specification

docs/skill-anatomy.md — Formal format spec for all BA-Kit skills. Every new contributor must follow this:

• Frontmatter (name, description, version)
• AGENCY block (role, tone, capabilities, goal)
• MEMORY block (required context)
• When to Use / NOT to Use
• System Instructions with System 2 Reflection
• Squad Handoffs
• Common Rationalizations (required)
• Red Flags (required)
• Verification (required)
• Knowledge Reference

5 Flagship Agents Updated

Applied the anti-rationalization pattern to the 5 most-used agents:

• @ba-master — Dispatcher rationalizations
• @ba-elicitation — Interview rationalizations
• @ba-writing — User Story rationalizations
• @ba-validation — QA rationalizations
• @ba-questioning — Interview prep rationalizations

Other 28 agents will receive the pattern in future incremental updates.

Stats

Metric	v3.1.1	v3.2.0	Change
Agents	33	33	-
Meta skills	0	1	+1
Slash commands	0	6	+6
Agents with anti-rationalization	0	5	+5
Docs files	13	14	+1 (skill-anatomy.md)

Philosophy

"Skills are processes, not prose. Every skill encodes hard-won engineering judgment as step-by-step workflows with anti-slack mechanisms."

Patterns inspired by production-grade engineering skills for AI coding agents.

Breaking Changes

None. All changes are additive.

Migration

No migration needed. Existing agent invocations continue to work. New sections are purely additive enforcement mechanisms.

BA-Kit v3.1.1 — License & Documentation

BA-Kit v3.1.1

License Change: MIT → CC BY-NC 4.0

BA-Kit is now licensed under Creative Commons Attribution-NonCommercial 4.0 International.

	MIT (old)	CC BY-NC 4.0 (new)
Personal use	✅	✅
Educational use	✅	✅
Commercial use	✅	❌ Requires separate license
Attribution	Optional	Required
Modification	✅	✅ (must share alike)

Why the change: BA-Kit contains 831 curated knowledge entries, 33 agent SKILL definitions, and synthesized book knowledge that represent significant intellectual effort. CC BY-NC protects this while keeping it free for the BA community.

Full license: CC BY-NC 4.0

---

Documentation Overhaul (13 files rewritten)

Complete rewrite of all documentation for new-user clarity and consistency:

File	Lines	What changed
docs/README.md	88	New onboarding navigation guide with ASCII reading-path flowchart
docs/prompt-library.md	523	48 prompts across 10 lifecycle phases (was duplicated/scattered)
docs/workflow-cookbook.md	295	15 clean scenarios (consolidated from 23 duplicated ones)
docs/quick-start.md	109	3-platform install, first use examples, FAQ
docs/usage-guide.md	157	Squad model, 3 workflow chains, tips
docs/junior-start.md	144	4-week onboarding with 7 new agents included
docs/contributing.md	104	New agent guide, CSV format, PR checklist
docs/agent-cheat-sheet.md	156	33 agents, power combos, decision matrix
docs/architecture-decisions.md	158	3 new ADRs (questioning, diagram, outputs)
docs/antigravity-protocol.md	165	v3.1.0, 33 agents roster, diagram mandate
docs/ai-foundation-for-ba.md	516	Updated counts (bilingual primer)
docs/ai-tools-guide.md	215	Verified accurate
docs/design-prototype-guide.md	260	@Ba-ux + @ba-diagram integration

Fixes

• All stale agent counts (19/21/26) eliminated
• All broken links in docs/README.md resolved
• Scenario count: 23 → 15 (removed duplicates)
• Prompt count: standardized to 48

Stats

Metric	Value
Agents	33
Knowledge Entries	831 (23 domains)
Templates	14
Prompts	48
Workflow Scenarios	15
Documentation Files	13
Total Doc Lines	2,880

BA-Kit v3.1.0 — The Lifecycle Release

BA-Kit v3.1.0 — The Lifecycle Release

33 Agent Specialists for Requirements Engineering with System 2 Reflection.

Highlights

Metric	v3.0.0	v3.1.0	Change
Agents	26	33	+7 new
Prompts	33	48	+15
Knowledge Entries	831	831	—
Domains	23	23	—
Templates	14	14	—
Diagram Types	—	24+	New

---

7 New Lifecycle Agents

Agent	Role	Key Capability
@ba-questioning	Master Interviewer	Paul-Elder Critical Thinking, 8 Elements of Reasoning, Intellectual Standards, Assumption Surfacing
@ba-communication	Communicator	Audience-adapted status reports, executive summaries, Pyramid Principle
@ba-ux	UX Researcher	Persona, Journey Map, Empathy Map, JTBD, UX Psychology, Heuristic Evaluation
@ba-data	Data Architect	ERD (Mermaid), Data Dictionary, DFD, Data Mapping, Migration Planning
@ba-change	Change Agent	ADKAR Assessment, Training Needs, Go-Live Checklist, Benefits Realization
@ba-business-rules	Rules Engine	Decision Tables, Decision Trees, Rule Catalog, Conflict Detection
@ba-diagram	Visualizer	Mermaid v11 (24+ diagram types), BA artifact→diagram mapping, Confluence export

---

Major Enhancements

@ba-questioning v2.1 — Complete critical thinking framework:

• 3 Question Types (One-System / No-System / Conflicting-System)
• 8 Elements of Reasoning (Purpose, Question, Information, Inferences, Concepts, Assumptions, Implications, Viewpoints)
• 8 Intellectual Standards audit (Clarity, Precision, Accuracy, Relevance, Depth, Breadth, Logic, Fairness)
• Wilson Method for concept disambiguation (Model / Contrary / Related / Borderline cases)
• Socratic Questioning Protocol (6-step deep probing)
• Prior Questions decomposition
• Interdisciplinary Domain decomposition
• Decision-Making & Problem-Solving question frameworks
• Bias Detection (Egocentric, Sociocentric, Stakeholder)
• 7 Intellectual Virtues for BA self-development
• Source: The Art of Asking Essential Questions (Elder & Paul)

@Ba-ux v2.0 — UX Psychology for BA:

• Cognitive Load Theory (progressive disclosure, sensible defaults, recognition over recall)
• Decision Architecture (default bias, choice paralysis, commitment escalation, loss aversion)
• Key UX Laws (Hick's, Fitts's, Jakob's, Peak-End, Gestalt)
• Information Architecture principles (navigation, F/Z patterns, all-states design)
• Usability Test Protocol template
• UX Heuristic Review Checklist (11 UX + 6 accessibility)
• Quick Validation Methods (5-second test, think-aloud, task completion)
• Source: legacy ux-designer skill (patterns-and-flows, psychology-deep-dive)

@ba-diagram — Confluence-ready Mermaid diagramming:

• 25 BA artifact types mapped to optimal Mermaid diagram types
• 8 ready-to-use BA diagram templates (BPMN swimlanes, stakeholder quadrant, approval sequence, user journey, sprint gantt, RTM chain, fishbone mindmap, sankey data flow)
• 4 Confluence embedding methods (Stratus mermaid-macro for DC, Cloud native, pre-render SVG, Draw.io)
• Hard rules for Confluence DC compatibility (discovered through production testing)

@ba-agile — Estimation Facilitation:

• Planning Poker, T-shirt Sizing, Story Point Calibration
• Story Splitting Patterns (6 patterns)
• Estimation Anti-Patterns BA should flag

@ba-wiki — Glossary Management:

• Ubiquitous Language enforcement
• Synonym/Homonym detection
• Term consistency checking across artifacts

@ba-confluence — Mermaid Handling:

• Stratus mermaid-macro plugin support for Confluence DC
• Code block language whitelist (safe vs broken)
• Validation via body.view (not body.storage)

---

Documentation Overhaul

Complete rewrite of all documentation for consistency and accuracy:

• README.md/vi.md: full rewrite (8 badges, ASCII flow, <details> agents)
• CLAUDE.md: new project root guidance for AI agents
• AGENTS.md: rewrite for OpenCode/Gemini compatibility
• docs/README.md: fixed all broken links
• docs/agent-cheat-sheet.md: 33 agents with power combos + decision matrix

Infrastructure

• outputs/ removed from git tracking (project deliverables are per-instance)
• .gitignore reorganized with Python artifacts, Claude session state, generated diagrams
• .claude/session-state/ cleaned from tracking

---

Platforms

Platform	How to Install
Antigravity IDE	cp -r .agent/skills/* ~/.gemini/antigravity/skills/
Claude Code	Clone repo, skills auto-detected from .agent/skills/
Claude CoWork	Upload individual SKILL.md files as project knowledge

Full Agent Roster (33)

@ba-master · @ba-identity · @ba-elicitation · @ba-writing · @ba-validation · @ba-traceability · @ba-nfr · @ba-process · @ba-prioritization · @ba-solution · @ba-conflict · @ba-export · @ba-metrics · @ba-root-cause · @ba-innovation · @ba-strategy · @ba-facilitation · @ba-systems · @ba-agile · @ba-test-gen · @ba-quality-gate · @ba-consistency · @ba-auditor · @ba-questioning · @ba-communication · @ba-ux · @ba-data · @ba-change · @ba-business-rules · @ba-diagram · @ba-jira · @ba-confluence · @ba-wiki

v2.9.3 — Standards & Completeness Update

What's New

3 Critical Gaps Closed

• PRD Template (templates/prd-template.md) — 12-section Product Requirements Document with OKR-native goals, MoSCoW features, AI-specific NFR section
• Design & Prototype Guide (docs/design-prototype-guide.md) — Stitch MCP workflows, Figma MCP setup, v0/Lovable/Bolt vibe-coding pipeline
• Prompt Library (docs/prompt-library.md) — 28 copy-paste prompts organized by 6 BA workflow phases

Document Reorganization

• All filenames standardized to kebab-case (14 renames)
• 3 overlapping architecture docs merged → docs/architecture-decisions.md
• New navigation index: docs/README.md
• Template headers cleaned (removed legacy SKILL-XX prefixes)

Quality

• Zero AI-generated marketing filler remaining
• All cross-references updated to renamed filenames
• Version badges synced to 2.9.3

New Cookbook Scenarios

• Scenario 21: Stitch MCP UI Pipeline
• Scenario 22: Design System Alignment
• Scenario 23: PRD-Driven Development

---

Full changelog: CHANGELOG.md

v2.5.0: The Antigravity Squad Protocol

🚀 Release 2.5.0: The Antigravity Squad Protocol

Date: 2026-01-07
Codename: "The Specialist Squad"

We have evolved. No longer a "Swarm" of bots, we are now a Cognitive Squad of 15 Specialists.
This release brings Military-Grade Professionalism and High-Assurance Capabilities to your workflow.

---

🌟 What's New?

1. ⚔️ The "Squad" Terminology Pivot

We have deprecated the biological "Swarm" metaphor in favor of the professional "Squad" (Biệt Đội) concept.

• Old: Insects, Instincts, Swarm Intelligence.
• New: Specialists, Collaboration, Mission Control.
• Why: To align with our System 2 (Critical Thinking) architecture. We don't just buzz; we think.

2. 📒 The Continuity Ledger (Shared Memory)

The Squad now shares a "Working Brain" via templates/CONTINUITY.md.

• Context Injection: Define your Goal & Constraints once.
• Squad Synchronization: All 15 agents read this file before acting.
• Result: Zero repetition. Maximum alignment.

3. 🛡️ "Level 5 Enabler" Maturity

We have refined our claim from "Level 5 Organization" to "High-Assurance Enabler".

• We provide the Tools (SPC, Root Cause, Python).
• You provide the Agency.
• Together, we achieve CMMI Level 5 behaviors.

---

🚀 Release 2.4.0: The Antigravity Native Update

Date: 2026-01-07
Codename: "System 2 Swarm"

We are proud to announce the biggest architectural shift in the history of BA-Kit.
We have moved beyond "Python Scripts" to a true Cognitive Swarm powered by the Antigravity Native Protocol (ANP).

---

🌟 What's New?

1. 🧠 System 2 Intelligence (The Reflective Loop)

Agents no longer just "output text". They now follow a strict cognitive process:

• System 1: Draft the answer.
• System 2: "Stop & Think" - Critique the draft for errors, hallucinations, or vague logic.
• Output: The verified result.

2. 🛡️ Tool Mandates (Hardened Reliability)

We have eliminated common LLM errors by forcing agents to use tools:

• Math: @ba-solution and @ba-metrics MUST use Python. No more mental math errors.
• Traceability: @ba-traceability MUST use grep to verify links. No more hallucinated dependencies.
• Standards: @ba-nfr MUST use search_web to verify ISO/GDPR clauses.

3. 🤖 The Full 15-Agent Roster

We have unlocked the "Level 5" capabilities with 3 new agents:

• @ba-metrics: Statistical Process Control (SPC) expert.
• @ba-root-cause: Deep investigator (5 Whys).
• @ba-innovation: R&D scientist (A/B Testing).

4. 📖 The Workflow Cookbook

A new guide WORKFLOW-COOKBOOK.md containing 10 Battle-Tested Scenarios (Startup, Enterprise, Crisis, etc.) to help you chain agents together for complex problem solving.

---

📦 Migration Guide

If you are coming from v1.x or v2.0 (Python Script era):

1. Stop running python ba-agent.py.
2. Copy the workflows:

   cp -r ba-kit/.agent/workflows/ ~/.gemini/antigravity/workflows/

3. Summon via Chat:

   "Hi @ba-identity, let's start a project."

---

🏆 Credits

Developed with the Super Ultra Deep Reasoning methodology.

• Architecture: Antigravity Native Protocol v2.4
• Compliance: CMMI Level 5, ISO 25010, IEEE 29148.

Code Less. Think More.

v2.4.0: Antigravity Native Swarm (System 2)

🚀 Release 2.4.0: The Antigravity Native Update

Date: 2026-01-07
Codename: "System 2 Swarm"

We are proud to announce the biggest architectural shift in the history of BA-Kit.
We have moved beyond "Python Scripts" to a true Cognitive Swarm powered by the Antigravity Native Protocol (ANP).

---

🌟 What's New?

1. 🧠 System 2 Intelligence (The Reflective Loop)

Agents no longer just "output text". They now follow a strict cognitive process:

• System 1: Draft the answer.
• System 2: "Stop & Think" - Critique the draft for errors, hallucinations, or vague logic.
• Output: The verified result.

2. 🛡️ Tool Mandates (Hardened Reliability)

We have eliminated common LLM errors by forcing agents to use tools:

• Math: @ba-solution and @ba-metrics MUST use Python. No more mental math errors.
• Traceability: @ba-traceability MUST use grep to verify links. No more hallucinated dependencies.
• Standards: @ba-nfr MUST use search_web to verify ISO/GDPR clauses.

3. 🤖 The Full 15-Agent Roster

We have unlocked the "Level 5" capabilities with 3 new agents:

• @ba-metrics: Statistical Process Control (SPC) expert.
• @ba-root-cause: Deep investigator (5 Whys).
• @ba-innovation: R&D scientist (A/B Testing).

4. 📖 The Workflow Cookbook

A new guide WORKFLOW-COOKBOOK.md containing 10 Battle-Tested Scenarios (Startup, Enterprise, Crisis, etc.) to help you chain agents together for complex problem solving.

---

📦 Migration Guide

If you are coming from v1.x or v2.0 (Python Script era):

1. Stop running python ba-agent.py.
2. Copy the workflows:

   cp -r ba-kit/.agent/workflows/ ~/.gemini/antigravity/workflows/

3. Summon via Chat:

   "Hi @ba-identity, let's start a project."

---

🏆 Credits

Developed with the Super Ultra Deep Reasoning methodology.

• Architecture: Antigravity Native Protocol v2.4
• Compliance: CMMI Level 5, ISO 25010, IEEE 29148.

Code Less. Think More.

💎 BA-Kit v2.2.0 - The Professional Branding Update

💎 Release Notes: v.2.2.0 - The Professional Branding Update

Released on 2026-01-04

🌟 Executive Summary

BA-Kit v2.2.0 marks the completion of our visual identity overhaul and repository optimization. With a new "World-Class" transparent logo (The Knowledge Block) and refined documentation structure, the kit effectively communicates its Antigravity CMMI Level 5 status.

We also introduced Context-Aware Best Practices, empowering AI Agents to navigate project-specific templates intelligently.

---

🎨 Visual Identity Upgrade

• New Logo: A minimalist, architectural "B" block in Midnight Blue & Metallic Gold.
• Improved Compatibility: Fully transparent background for seamless integration in Dark/Light themes.
• Coverage: Applied across all documentation headers and assets.

🧹 Repository Hygiene

• Cleanup: Eliminated broken links and deprecated file references (old 00-MASTER-INDEX.md).
• Standardization: Unified README and QUICK-START structures.

🚀 Upgrade Guidance

No breaking changes. Users can verify the new branding by checking the header of any README or USAGE-GUIDE.

For AI Agent integration:

• Review the new "Context-Aware Template Organization" section in USAGE-GUIDE.vi.md to optimize your project folders.

---

BA-Kit Team | Antigravity

🚀 BA-Kit v2.1.0 - The Experience Update

🚀 BA-Kit v2.1.0 - The Experience Update

Persona Guide, Workflow Optimization & Repository Cleanup

Release Date: 2026-01-04

---

🏆 Executive Summary

Following the massive CMMI Level 5 update (v2.0), this v2.1.0 "Experience Update" focuses on usability and adoption. We've added a comprehensive Persona-Based User Guide for different experience levels, streamlined the repository by removing redundant files, and fully updated all agent workflows to utilize the new 20-skill framework.

---

⭐ Highlights

1. 🆕 Persona-Based User Guide

File: USER-GUIDE-BY-PERSONA.md

A massive new bilingual (English-Vietnamese) guide that teaches BA-Kit through a real-world Attendance System project:

Level	Experience	Focus	Output
🟢 Beginner	1-2 years	Doing the work right	User Stories, Use Cases
🟡 Intermediate	3-5 years	Doing the right work	NFRs, Data Models, Prioritization
🔵 Advanced	5+ years	Improving how work is done	ROI, SPC, Root Cause Analysis

2. 🔄 Workflow & Agent Updates

• Master Workflow: ba-master.md updated with:
  • Pattern F: Process Improvement (CMMI Level 5)
  • Pattern G: Business Case Development
  • New Advanced Skills section (SKILL-18, 19, 20)
• Agent Awareness: AGENT.MD updated with 8 new triggers for advanced skills.

3. 🧹 Repository Purification

• Removed 00-MASTER-INDEX.md: Content was redundant with README and QUICK-START.
• Simplification: Reduced root-level files to clear confusion ("Which file do I read first?").

---

📋 Changelog

Added

• USER-GUIDE-BY-PERSONA.md: 600+ lines of guide content.
• New Patterns: Process Improvement and Business Case patterns in ba-master.md.

Changed

• AGENT.MD: Updated to 20 skills, added Six Sigma cert, improved trigger table.
• QUICK-START.md: Updated to 20 skills, 4 categories.
• ba-master.md: Added CMMI Level 5 badge and advanced workflows.

Removed

• 00-MASTER-INDEX.md: Deleted to improve navigation clarity.

---

🚀 Upgrade Guidance

• For New Users: Start with README.md then jump to USER-GUIDE-BY-PERSONA.md to find your level.
• For AI Agents: Re-read AGENT.MD to update your internal context window with the new triggers.

---

BA-Kit v2.1.0 — Making World-Class Business Analysis Accessible to Everyone