bkit v1.5.8 — PM Agent Team, Skills 2.0 & Team Orchestration

Highlights

bkit v1.5.8 is the largest feature release since v1.5.0, introducing the PM Agent Team for product discovery, Skills 2.0 Classification, and a complete Team Orchestration system with 5 coordination patterns.

PM Agent Team — Product Discovery Before Code

5 new specialized PM agents enable structured product analysis before writing a single line of code:

Agent	Role
pm-lead	Orchestrates 4-phase discovery workflow
pm-discovery	Opportunity Solution Tree (Teresa Torres)
pm-strategy	Value Proposition (JTBD) + Lean Canvas
pm-research	Personas, competitor analysis, market sizing (TAM/SAM/SOM)
pm-prd	8-section PRD synthesis with beachhead segment & GTM strategy

Run /pdca pm <feature> to generate a comprehensive PRD before planning.

Skills 2.0 — Smarter Skill Activation

All 35 skills now carry a classification tag — Workflow (W), Capability (C), or Hybrid (H) — enabling the skill orchestrator to make better activation decisions. 6 new skills added:

• /plan-plus — Brainstorming-enhanced planning with intent discovery and alternatives exploration
• /simplify — Code quality review and complexity reduction
• /batch — Process multiple features through PDCA in parallel
• /loop — Run commands on recurring intervals (e.g., /loop 5m /pdca status)
• /output-style-setup — Install output style files
• /pm-discovery — PM Agent Team product discovery workflow

Team Orchestration — Coordinated Multi-Agent Workflows

9 new modules in lib/team/ with 5 coordination patterns for Dynamic and Enterprise projects:

Pattern	Use Case
Leader	Standard feature development (1 lead + N workers)
Council	Architecture decisions (equal peer review)
Swarm	High-parallelism tasks (dynamic worker pool)
Pipeline	Multi-phase sequential workflows
Watchdog	Continuous monitoring with automated responses

Infrastructure

• Path Registry (lib/core/paths.js) — Centralized state file management with auto-migration from legacy paths
• Language Patterns (lib/intent/language-patterns.js) — Structured 8-language keyword patterns for intent detection
• 7th @import Module — executive-summary-rules.md for mandatory summary output after PDCA document work

---

What Changes for Users

Before v1.5.8

• Planning started immediately with /pdca plan — no structured product analysis
• Skills were flat — no classification, no activation intelligence
• Team mode was basic — limited coordination patterns
• 16 agents, 29 skills, 18 commands

After v1.5.8

• Product discovery first: /pdca pm <feature> generates a comprehensive PRD with personas, market analysis, and GTM strategy — then /pdca plan auto-references it
• Smarter skill suggestions: Skills are classified (W/C/H) so the system recommends the right skill for the right context
• Parallel team workflows: 5 orchestration patterns let multiple agents coordinate on complex features
• Enhanced planning: /plan-plus adds brainstorming phases (intent discovery, alternatives, YAGNI review) before document generation
• Automation tools: /loop for recurring monitoring, /batch for parallel multi-feature processing, /simplify for code quality passes
• 21 agents, 35 skills, 24 commands — significantly expanded capabilities

---

Component Counts

Category	v1.5.7	v1.5.8	Change
Agents	16	21	+5
Skills	29	35	+6
Commands	18	24	+6
Test Suites	24	78	+54
Test Cases	~218	972	+754
Lib Modules	~12	~25	+13
@import Modules	6	7	+1

Compatibility

• Gemini CLI: v0.29.0 ~ v0.33.x (tested through v0.33.0-preview.4)
• Claude Code: v2.1.34 ~ v2.1.71 (recommended: v2.1.71+)

Quality

• 972 / 972 tests passing (100%) across 78 test suites
• 11-perspective test coverage: unit, integration, E2E, scenario, philosophy, security, edge cases, boundary, recovery, compatibility, template
• 3 PDCA cycles completed at 100% match rate (implementation, testing, documentation)

---

Full Changelog: v1.5.7...v1.5.8

bkit v1.5.7 - Gemini CLI v0.32.x Full Migration

Highlights

bkit v1.5.7 completes the full migration to Gemini CLI v0.32.x with significant new capabilities across tools, hooks, policies, and testing infrastructure.

New Features

• 6 Task Tracker Tools: read_task_data, update_task_data, read_many_task_data, update_many_task_data, list_tasks, add_task_to_session — first-party PDCA integration with Gemini CLI's task system
• SDK RuntimeHook Dual-Mode: 6 hot-path hooks (before_model, after_model, before_tool, after_tool, before_agent, after_agent) now support both legacy shell and SDK runtime modes
• 4-Tier Policy Engine: bkit-starter-policy.toml + policy-migrator.js with auto-detection for Starter/Dynamic/Enterprise/Custom tiers
• Task Tracker - PDCA Bridge: tracker-bridge.js module synchronizes Gemini task tracker state with bkit PDCA status
• 11 New Feature Flags for v0.32.0+ (29 total for v0.32.1)

Infrastructure

• 4 New Test Suites: TC-21 (v0.32 Migration), TC-22 (PDCA Status Path), TC-23 (Tracker Bridge), TC-24 (Runtime Hooks)
• 207 Tests Passing (100% pass rate)
• Gemini CLI Support: v0.29.0 ~ v0.32.1

Documentation

• Complete version synchronization across all config files, hook scripts, and lib modules
• Archived gemini-cli-032-migration PDCA documents
• Full PDCA cycle completed with 100% match rate

Stats

Metric	Value
Built-in Tools	23 (was 17)
Agents	16 (4 with tracker)
Skills	29 (3 with tracker)
Hook Events	10 (6 SDK dual-mode)
Test Suites	TC-01 ~ TC-24
Policy Tiers	4
Feature Flags	29

---

Full Changelog: v1.5.6...v1.5.7

bkit v1.5.6 - Gemini CLI v0.31.0 Compatibility

Highlights

Gemini CLI v0.31.0 Full Compatibility

bkit now supports 9 new feature flags from Gemini CLI v0.31.0, including RuntimeHook functions, Browser Agent, Tool Annotations, Project-Level Policy, MCP Progress, Parallel Read Calls, and more. The version detector automatically enables the right capabilities based on your installed CLI version.

Tool Annotations for All 17 Built-in Tools

Every Gemini CLI tool now has readOnlyHint, destructiveHint, and idempotentHint metadata annotations. This enables the CLI's trust model to make smarter decisions about tool execution and parallel optimization — tools like read_file and grep_search are marked read-only, while run_shell_command is correctly flagged as destructive.

Level-Specific Policy Engine

bkit automatically generates Tier 3 TOML policy files tailored to your project level:

• Starter: Restrictive defaults — write/shell commands require confirmation, dangerous operations denied
• Dynamic: Balanced — standard operations allowed, destructive commands blocked
• Enterprise: Permissive — full access with audit trail for force-push operations

Hook Adapter Module (v1.6.0 Migration Prep)

New hook-adapter.js module detects RuntimeHook function support and prepares the SDK event mapping (PascalCase → snake_case) for the upcoming v1.6.0 migration from command-based hooks to in-process SDK hooks (99% faster execution).

Enhanced Version Detection

Feature flags are now organized by CLI version milestone (v0.26.0+, v0.29.0+, v0.30.0+, v0.31.0+) with 17 total flags, making it clear which capabilities are available at each version level.

---

What Changed for Users

If you're on Gemini CLI v0.31.0+

• Automatic policy generation — bkit creates project-level TOML policies matching your detected level (Starter/Dynamic/Enterprise) on first session start. No manual configuration needed.
• Smarter tool execution — Tool annotations enable the CLI to auto-approve read-only operations and flag destructive ones, reducing unnecessary permission prompts.
• Session metadata enriched — Your session start now reports CLI version, feature flag counts, and preview status for better diagnostics.

If you're on Gemini CLI v0.29.0–v0.30.0

• No breaking changes — All existing functionality works exactly as before. The new features gracefully degrade when the CLI doesn't support them.
• Forward compatibility — When you upgrade to v0.31.0+, bkit will automatically enable the new capabilities without any configuration changes.

For all users

• Cleaner documentation — 60+ legacy PDCA documents archived, leaving only active v1.5.6 content in the docs directory. Faster context loading and reduced noise.
• 3 new test suites (TC-18, TC-19, TC-20) covering Tool Annotations, Level Policy, Hook Adapter, and coverage gaps — total test coverage expanded significantly.

---

Added

• lib/adapters/gemini/hook-adapter.js — RuntimeHook function detection + SDK event name mapping (84 lines)
• TOOL_ANNOTATIONS — readOnlyHint, destructiveHint, idempotentHint for all 17 built-in tools
• getToolAnnotations(), isReadOnlyTool(), getStrictReadOnlyTools() query functions
• 9 new v0.31.0+ feature flags: hasRuntimeHookFunctions, hasBrowserAgent, hasProjectLevelPolicy, hasMcpProgress, hasParallelReadCalls, hasPlanModeCustomStorage, hasToolAnnotations, hasExtensionFolderTrust, hasAllowMultipleReplace
• LEVEL_POLICY_TEMPLATES + generateLevelPolicy() for Starter/Dynamic/Enterprise
• getGeminiCliFeatures() in session-start.js metadata
• 3 test suites: tc18-v031-features.js, tc19-v031-policy-hooks.js, tc20-coverage-gaps.js
• .gemini/policies/bkit-permissions.toml base policy file
• Tool Annotations section in .gemini/context/tool-reference.md

Changed

• Feature flags organized by CLI version milestone (v0.26.0+ through v0.31.0+)
• Session start generates level-specific policies when hasProjectLevelPolicy is true
• Test runner includes TC-18/19/20, report path updated to v1.5.6
• TC-04 permission tests use isolated test projects
• 0.31.0 added to bkit.config.json testedVersions
• All documentation synchronized to v1.5.6 (README, GEMINI.md, CHANGELOG, guides)
• 60+ legacy PDCA documents archived to docs/archive/2026-02/
• PDCA status cleaned: 15 features → 3 active v1.5.6 features

Compatibility

Requirement	Version
Gemini CLI	v0.29.0+ (forward-compatible through v0.31.0)
Node.js	v18+

Full Changelog: v1.5.5...v1.5.6

bkit-gemini v1.5.5 - Gemini CLI 0.3.0 Migration & Security Hardening

Highlights

🔒 Critical Security Fixes

• Fixed unconditional --yolo flag in sub-agent spawning that bypassed all safety prompts
• Fixed team_name path traversal vulnerability in team_create handler
• Added SemVer format validation to block env var injection (GEMINI_CLI_VERSION=99.99.99)
• Added TOML string escaping to prevent policy injection via backslash, quote, and newline characters

🚀 Gemini CLI v0.3.0 Compatibility

• Policy TOML Auto-Generation: session-start.js triggers policy-migrator.js when Gemini CLI >= v0.30.0 is detected, creating .gemini/policies/bkit-permissions.toml automatically
• Version-Aware Approval Flag: spawn-agent-server.js uses --approval-mode=yolo for v0.30.0+ and --yolo for older versions
• TOML Structural Validation: Validates generated TOML before writing (rule count vs decision count match)
• New Feature Flags: hasGemini31Pro (v0.29.7+), hasApprovalMode (v0.30.0+)

🧠 Model Upgrades

• cto-lead & gap-detector: Upgraded to gemini-3.1-pro (ARC-AGI-2 77.1%)
• report-generator & qa-monitor: Switched to gemini-3-flash-lite (60% cost reduction)

🛡️ Enhanced Safety

• before-tool.js: Now blocks reverse shells, policy file tampering, remote code execution via pipes, and sensitive file access
• after-tool.js: Defensive field access supports both tool_name/toolName and tool_input/toolInput variants

🧪 Test Infrastructure

• tc16-v030-phase1.js: Version detection and feature flag validation tests
• tc17-v030-phase2.js: Policy migration and tool registry tests
• Manual smoke test checklist: tests/manual/phase1-smoke-checklist.md

📚 Documentation

• model-selection.md: Added Gemini 3.1 Pro + customtools variant documentation
• SDK Dependency Integration Research: Comprehensive analysis of SDK integration patterns
• Full PDCA Cycle: Plan, Design, Analysis, and Report documents for both gemini-cli-030-migration and bkit-v155-gemini-test features

---

Stats

• 40 files changed (+7,768 / -96)
• 4 critical/high security fixes
• 2 new test suites + manual checklist
• 14 new documentation files

Full Changelog: v1.5.4...v1.5.5

v1.5.4 - Gemini 3 Model Migration & Policy Engine Preparation

Highlights

Gemini 3 Model Migration — All 16 bkit agents upgraded from gemini-2.5-pro/flash to gemini-3-pro/flash with optimized temperature settings (minimum +0.1 to prevent looping behavior).

Version Detector Module — New version-detector.js provides 3-strategy Gemini CLI version detection (environment variable → npm registry → CLI execution) with feature flags and 5-minute caching.

Policy Engine Preparation — New policy-migrator.js converts bkit.config.json permissions to Gemini CLI v0.30.0 Policy Engine TOML format, ensuring forward compatibility.

Forward Alias Layer — Tool registry now includes FORWARD_ALIASES mapping for upcoming tool name changes (replace→edit_file, glob→find_files, grep_search→search_files, etc.).

---

What's Changed

Added

• Version Detector Module: lib/adapters/gemini/version-detector.js — 3-strategy fallback (env var, npm, CLI), feature flags, caching
• Policy Migrator Module: lib/adapters/gemini/policy-migrator.js — bkit.config.json → Policy Engine TOML converter
• Forward Alias Layer: FORWARD_ALIASES in tool-registry.js for future tool name changes
• Compatibility Config: bkit.config.json new compatibility section with version tracking and Policy Engine settings
• Agent Memory Storage: .gemini/agent-memory/bkit/ for all 16 agents
• 5 New Test Suites: tc11 (output-styles), tc12 (agent-memory), tc13 (automation), tc14 (bkend-skills), tc15 (feature-report)
• PDCA Documentation: Complete Plan, Design, Analysis, and Report documents for v1.5.4 features

Changed

• 16 Agents Model Update: gemini-2.5-pro → gemini-3-pro (9 agents), gemini-2.5-flash → gemini-3-flash (7 agents)
• 16 Agents Temperature Optimization: Adjusted for Gemini 3 recommended ranges
• gemini-extension.json: Removed deprecated excludeTools field (v0.30.0 Policy Engine migration)
• GeminiAdapter Version: 1.0.0 → 1.5.4 with getCliVersion() and getFeatureFlags() methods
• Permission Manager: Added Policy Engine fallback — defers to native TOML policies when .gemini/policies/*.toml detected

Documentation

• tool-reference.md: Added Forward Aliases table for future tool naming compatibility
• README.md: Updated version badges and compatibility section for v1.5.4
• CHANGELOG.md: Full v1.5.4 release notes

Compatibility

• Minimum: Gemini CLI v0.29.0 (unchanged)
• Forward-compatible with Gemini CLI v0.30.0 Policy Engine

---

Full Changelog: v1.5.3...v1.5.4

v1.5.3 - Gemini CLI v0.29.0+ Compatibility

Gemini CLI v0.29.0+ Compatibility Fix

This release fixes all invalid tool name errors (Issue #5) that occurred when running bkit-gemini on Gemini CLI v0.29.0+.

Highlights

• Tool Registry — New centralized module (lib/adapters/gemini/tool-registry.js) serving as the single source of truth for all 17 built-in Gemini CLI tool names
• 61 files fixed — All 16 agents, 29 skills, hook scripts, and adapter layer updated with correct tool names
• Backward compatible — Legacy tool name aliases (glob_tool, web_search, task_write) are resolved automatically via the registry

Breaking Changes

• Minimum Gemini CLI version: v0.29.0 (was v0.28.0)
• Removed experimental block from gemini-extension.json (Skills/Hooks GA since v0.26.0)

Tool Name Changes

Old (Invalid)	New (Correct)	Scope
glob_tool	glob	16 agents, 29 skills
web_search	google_web_search	7 agents, 10 skills
spawn_agent	(removed — not a built-in tool)	1 agent, 1 skill
skill	activate_skill	hooks
grep	grep_search	adapter

Added

• lib/adapters/gemini/tool-registry.js — Centralized tool name registry with 17 verified names
• resolveToolName() — Legacy-to-current name resolution
• isValidToolName() — Validation API for frontmatter checks
• 4 new test cases for Tool Registry verification (CFG-03, CFG-06, CFG-07, CFG-08)

Fixed

• All 16 agent frontmatter tools: arrays
• All 29 skill frontmatter allowed-tools: arrays
• Hook scripts (before-tool-selection.js, after-tool.js)
• Hook config matcher in hooks.json
• Adapter TOOL_MAP in lib/adapters/gemini/index.js

Test Results

• Shell E2E: 42/42 (100%)
• Node.js Unit: 70/72 (97.2%) — 2 pre-existing E2E failures unrelated to this release

Full Changelog: v1.5.2...v1.5.3

bkit-gemini v1.5.2 - bkend.ai BaaS Integration

bkit-gemini v1.5.2 — bkend.ai BaaS Integration

AI-native development toolkit implementing PDCA methodology with Context Engineering for Gemini CLI

Highlights

🔗 bkend.ai Documentation Sync

Full integration of bkend.ai Backend-as-a-Service documentation into bkit skills. Developers can now build fullstack apps with authentication, database, storage, and more — all guided by AI agents without leaving the CLI.

5 New BaaS Skills (29 Total)

Skill	Purpose
bkend-quickstart	Platform onboarding, MCP setup, resource hierarchy
bkend-auth	Email/social login, JWT tokens, RBAC, RLS policies
bkend-data	Table CRUD, filtering, sorting, pagination, relations
bkend-storage	File upload/download, presigned URLs, CDN, bucket management
bkend-cookbook	10 single-project + 4 full-guide practical tutorials

28 MCP Tools (up from 6)

Expanded MCP tool registry with bkend.ai platform tools for direct backend operations — table management, authentication, file storage, and more — all accessible through natural language commands.

Updated Help Menu

/bkit command now displays the new Project Initialization & BaaS section with quick access to /bkend-quickstart, /bkend-auth, and /bkend-data skills.

v1.5.2 Full Test Verification

Comprehensive test suite with 68 test cases across all categories — 100% pass rate verified through PDCA gap analysis.

---

What's New

Added

• 5 new bkend.ai BaaS skills (29 total): quickstart, auth, data, storage, cookbook
• 28 MCP tools integrated from bkend.ai platform (up from 6)
• v1.5.2 full test plan and completion report
• github-stats.toml added to .gitignore for private commands

Updated

• commands/bkit.toml help menu: v1.5.1 → v1.5.2 with BaaS section
• PDCA status tracking for bkend-docs-sync and v1.5.2-full-test features
• Session memory and configuration metadata
• v1.5.1 test report consolidated

PDCA Cycle

• Plan: bkend-docs-sync feature scoped with 5 skill targets
• Design: Mapping of bkend.ai docs to skill structure
• Do: Implementation of all 5 skills with source documentation sync
• Check: Gap analysis — 100% match rate (3 findings resolved)
• Act: Completion report generated, merged to main

---

Quality Metrics

Metric	Result
Test Cases	68 scenarios (100% pass)
Gap Analysis	100% match rate
PDCA Iterations	1 (passed on first fix cycle)
New Skills	5 BaaS skills verified against source docs

Compatibility

• Gemini CLI: v0.28.0+
• Node.js: v18+ (for hook scripts)

Installation

gemini extensions install https://github.com/popup-studio-ai/bkit-gemini.git

Documentation

• README
• CHANGELOG

---

Full Changelog: v1.5.1...v1.5.2

bkit-gemini v1.5.1 - Context Engineering Enhancement

bkit-gemini v1.5.1 — Context Engineering Enhancement

AI-native development toolkit implementing PDCA methodology with Context Engineering for Gemini CLI

Highlights

16 Specialized Agents

5 new agents join the team: cto-lead, frontend-architect, security-architect, product-manager, qa-strategist — all with Gemini native frontmatter (model, tools, temperature, max_turns, timeout_mins).

10-Event Hook System

3 new lifecycle events: BeforeModel, AfterModel, BeforeToolSelection — providing full Gemini CLI lifecycle coverage from session start to end. 17 hook scripts total including 5 per-skill hooks and 2 utility modules.

Context Engineering Architecture

Three-layer architecture for systematic context curation:

Layer	Components	Count
Domain Knowledge	Skills	21
Behavioral Rules	Agents	16
State Management	Hook Scripts + Lib Modules	17 + 6

3 New Core Modules

Module	Lines	Purpose
Skill Orchestrator	708	Custom YAML parser, agent delegation, template auto-loading
Agent Memory	214	Per-agent persistent storage with project/user scope (max 20 sessions)
Context Hierarchy	209	4-level config merge: Plugin → User → Project → Session with TTL cache

4 Output Styles

• bkit-learning — Beginner-friendly step-by-step explanations
• bkit-pdca-guide — Standard PDCA workflow guidance
• bkit-enterprise — Enterprise-level technical focus
• bkit-pdca-enterprise — Combined enterprise and PDCA methodologies

Team Mode Foundation

3 new MCP tools for agent team coordination: team_create, team_assign, team_status — supporting dynamic (3 agents), enterprise (5 agents), and custom (up to 10 agents) strategies.

@import Modularization

GEMINI.md split into 6 focused context modules in .gemini/context/: pdca-rules, commands, agent-triggers, skill-triggers, tool-reference, feature-report.

8-Language Auto-Detection

Natural language triggers in English, Korean, Japanese, Chinese, Spanish, French, German, and Italian — automatically activating the appropriate agent or skill.

---

What's New (Full List)

Added

• 5 new agents (16 total) with Gemini native frontmatter
• 3 new hook events (10 total) for full lifecycle coverage
• 7 per-skill hook scripts + 2 utility modules (17 total)
• 4 output styles with level-based defaults
• Agent Memory system with project/user scope persistence
• Context Hierarchy with 4-level config merge and TTL cache
• Skill Orchestrator with custom YAML parser and agent delegation
• @import GEMINI.md modularization (6 context modules)
• Team Mode foundation with 3 MCP tools (6 total)
• Enhanced TOML commands with @{path}, !{command}, {{args}} syntax
• 10+ YAML frontmatter fields for all 21 skills
• 258 test cases across 10 categories (100% pass rate)
• CHANGELOG.md with full version history
• README.md complete rewrite with architecture docs

Enhanced

• Permission Manager with glob pattern matching and PDCA phase restrictions
• Context Fork with LRU(10), named snapshots, and diff capabilities
• Import Resolver with variable substitution and circular dependency detection
• SessionStart hook with dynamic context injection (392 lines)
• BeforeAgent hook with 8-language intent detection and ambiguity scoring
• bkit.config.json with 5 new sections
• MCP Server extended from 3 to 6 tools (753 lines)

---

Quality Metrics

Metric	Result
Test Cases	258 scenarios, 63 automated (100% pass)
Gap Analysis	135/135 verification points
PDCA Iterations	Passed on first check
Gemini CLI Compatibility	v0.28.0+

Compatibility

• Gemini CLI: v0.28.0+
• Node.js: v18+ (for hook scripts)

Installation

gemini extensions install https://github.com/popup-studio-ai/bkit-gemini.git

Documentation

• README
• CHANGELOG

---

Full Changelog: v1.5.0...v1.5.1

bkit v1.5.0 - Enhanced GEMINI.md and Documentation

bkit v1.5.0 Release

🎉 bkit Vibecoding Kit v1.5.0 - AI-native development toolkit for Gemini CLI

✨ Highlights

Enhanced GEMINI.md (Context Engineering)

Based on comprehensive research of official Gemini CLI documentation, GEMINI.md has been completely redesigned:

• 11 Agent Triggers with 8-language keyword support (EN, KO, JA, ZH, ES, FR, DE, IT)
• 13 Skill Triggers with auto-detection keywords
• Tool Name Reference table for Gemini CLI native tools
• Feature Usage Report format specification
• PDCA Phase Recommendations table

Documentation Improvements

• Added README images (smart-onboarding, how-to-use)
• Added "agentic", "agents", "workflow" keywords to extension manifest

Technical Updates

• Converted commands from .md to .toml format (Gemini CLI standard)
• Added test infrastructure with comprehensive verification scripts
• Added new library modules: memory.js, permission.js, dependency.js
• Added MCP spawn-agent server for sub-agent orchestration

📦 What's Included

Component	Count	Description
Skills	21	Domain-specific knowledge (phases, levels, specialized domains)
Agents	11	Role-based AI assistants with behavioral constraints
Commands	10	TOML-based slash commands
Hooks	7	Event-based automation (SessionStart → SessionEnd)
Library Modules	4	Core utilities (core, pdca, intent, task)

🚀 Installation

gemini extensions install https://github.com/popup-studio-ai/bkit-gemini.git

📝 Key Resources

• GEMINI.md - Enhanced context file with all triggers and rules
• README.md - Complete documentation with images
• Skills - 21 domain-specific skills
• Agents - 11 specialized agents

🔗 References

This release incorporates best practices from:

• Gemini CLI Official Documentation
• Gemini CLI Extensions Guide
• Agent Skills Documentation

---

Full Changelog: v1.0.0...v1.5.0

Made with AI by POPUP STUDIO