v11.0.0

11.0.0 (2026-04-18)

⚠ BREAKING CHANGES

• Artifact output path changed from docs/agent-system/ to .rcc/, and docs/validation-reports/ to .rcc/validation/. Existing projects are auto-migrated by migrating-agent-systems Task 2 on next run (git mv for tracked files, mv for untracked). Update any external tooling that referenced the old paths.
• All commands now use rcc: namespace
  • /reflect → /rcc:reflect
  • /init-project → /rcc:init-project
  • /migration → /rcc:migration
  • /refactor-skills → /rcc:refactor-skills
  • /improve-skill → /rcc:improve-skill
  • /add-law → /rcc:add-law
  • /create-plugin → /rcc:create-plugin

Features

• add /add-law command, bump to v4.5.0 (fdea049)
• Add /init-project command, bump to v4.0.0 (736b054)
• Add /init-project command, bump to v4.0.0 (8159b47)
• add /reflect command and update README with SDD workflow (7669103)
• add /reflect command and update README with SDD workflow (6b34059)
• add chrollo commands for external skill hunting, bump to v5.2.0 (31253e5)
• add command entries for new skill chain, split refactor command (ebabb87)
• add comprehensive best practices from write-skill to refactor-s… (440572d)
• add comprehensive best practices from write-skill to refactor-skills command (ffd32e2)
• add human-in-the-loop discovery to brainstorming workflow (16852f6)
• Add modular rules architecture and new skills (v3.0.0) (8900399)
• add project-discovery skill, bump to v5.1.0 (59480d4)
• add rcc: prefix to all commands, bump to v5.0.0 (bca24a9)
• add refactor-claude-md command with constitution mechanism (09efefc)
• add scripts for precision principle to Core Principles (774c383)
• add skill extraction guideline to refactor-claude-md command (d97fba7)
• add Skill Structure section and update line limit to 200 (e141c28)
• add write-command skill and refactor-skills command (ddbcd15)
• add write-skill and rename existing to write-plugin (34bcb22)
• analyzing: add pipeline state persistence checks to weakness checklist (41d310d)
• analyzing: add project scanning reference (750cabf)
• analyzing: add restructuring guide reference (a10a3a0)
• analyzing: expand report template with project overview and recommendations (60b4cc2)
• analyzing: restructure to 6 tasks with project scanning and recommendations (0f932ff)
• brainstorming: add targeted exploration questions reference (6d5dd0b)
• brainstorming: add targeted exploration questions reference (4714fda)
• brainstorming: redesign as targeted exploration with pipeline and pain point focus (5b8e016)
• brainstorming: update summary template for new task structure (0ced495)
• bump to v10.0.0 with advanced security architecture (b5514a8)
• bump to v9.8.0, enhance hook and agent validation (16992dc)
• enhance reflecting skill with agent correction triggers (d7f7f35)
• hooks: add batch validator + validating-plugins skill (fbe64ec)
• hooks: add frontmatter parser (882279c)
• hooks: frontmatter validator hook — complete implementation (c75537e)
• hooks: implement agent and rules frontmatter checks (8791c5e)
• hooks: implement plugin root discovery (7d4d04d)
• hooks: implement SKILL.md checks ①②③④ (d7ad8c8)
• hooks: register frontmatter validator hook (53fd08a)
• hooks: wire main entry point and plugin validate check (3ad79cb)
• implement AI-specialized security check patterns (f8d4caf)
• implement comprehensive security integration test suite for Tasks 1-4 (8c39ee0)
• implement long-term memory learning system for failure prevention (8d9b074)
• initialize claude code skill plugin repository (34f5839)
• integrate Qualys 2026 eight-layer defense architecture (9525677)
• Opus 4.7 alignment, safety_bypass, release-please, .rcc/config.yml (7755987)
• planning: read reflection reports in Task 1 inputs (e57c271)
• rcc: add Category 11 Rules Health to weakness checklist (8703ace)
• rcc: add content validation and red flags to writing-rules (0799722)
• rcc: add context and agent selection guide to writing-skills (2dc962a)
• rcc: add load cost, content classification, glob validity to rule-reviewer (ec258d7)
• rcc: add migrating-plugins skill, bump to v9.4.0 (461aefd)
• rcc: add model, isolation, and effort selection guide to writing-subagents (42a0053)
• rcc: add pre-plugin maturity level to migrating-plugins (8666cad)
• rcc: add project context completeness and user-root scope (v8.4.0) (2156a41)
• rcc: add Rules Health analysis to analyzing-agent-systems (e95b3a4)
• rcc: add Rules Health Summary to analysis report template (964f52b)
• rcc: add rules refactoring proposal to migrating-agent-systems (79de525)
• rcc: add script/template automation checks to reviewer, planner, refactorer (cc7d4fe)
• rcc: add templates/ as skill asset type alongside scripts/ and references/ (a8cddf9)
• rcc: replace writing-rules Task 1 decision tree with flowchart (929f8e9)
• rcc: restore init command for project initialization (f179a06)
• rcc: set reviewer agents to sonnet model with medium effort (713730d)
• rcc: update agent config and expand validator allowed fields (32d685d)
• refactoring: add Task 6 for pipeline document archiving (13e0d8d)
• reflecting skill 加入 consolidation review 並強化 router 追溯 (6e933c1)
• reflecting skill 加入 skill router 追溯機制 (fae4dc9)
• reflecting: add reflection report template (0690c2c)
• reflecting: rewrite as lightweight pipeline entry point (af615e6)
• v10.10.0 — anti-drift docs fetcher + bash hook fallback fix (0360ca3)
• v10.10.1 — anti-drift docs fetcher + bash hook fallback fix + writing-rules component fix (d468764)
• v10.8.0 — three-layer model architecture, YAML reviewer output, implementer capability analysis (8851c62)
• v10.9.0 — plugin health fixes, README, skill size reduction (ed1caee)
• v10.9.3 — hook and skill scripts use project primary language (3e7bcff)
• v7.0.0 - TDD-based skill design with task enforcement (b2ab488)
• v7.1.0 - align skills with Claude Code docs, add advising-architecture skill (55c6a61)
• v8.0.0 - decompose agent system workflow into 5-skill chain (4aef24f)
• v8.1.0 - add review step to core chain with 5 reviewer agents (5f0a5f1)
• v8.2.0 - add refactoring-plugins skill with cross-platform health check (20ab438)
• v8.3.0 - formalize routing patterns across all skills (236a293)
• v9.0.0 - align with agent engineering best practices (25bc21f)
• Windows hook compatibility, reflect pain-point arg, init-plugin command (05c734b)
• 實現 Agent 系統效能優化與護欄調整 (69cb189)

Bug Fixes

• add critical performance and timeout controls to AI security hook (7bad589)
• Add hooks and rules to migration command, bump to v3.1.0 (53ead42)
• Add missing write-rules and write-hook skills to ACE-core (7097c33)
• allow empty string matcher in hook validator, bump v10.9.1 (9a9fcd0)
• complete gerund naming and version bump to v6.0.0 (01a9a69)
• correct CLAUDE.md path in documentation (e4b2c37)
• correct GitHub username in README (ac6940c)
• correct marketplace source path format (c009435)
• enforce explicit model field, treat inherit as anti-pattern, bump v10.9.2 (fe813f3)
• enforce skill delegation in migration, init-project, and Law 2 (075fbbf)
• enforce skill delegation in migration, init-project, and Law 2 (54fd684)
• hooks: accept markdown link OR text mention for all file types (9bdba7a)
• hooks: add python3 detection with user feedback, bump v9.1.1 (be52945)
• hooks: correct orphaned check for scripts vs .md references (62d1008)
• hooks: normalize linked paths in orphan detection (afbbb75)
• hooks: remove uv fallback that consumed stdin (f87d404)
• hooks: use additionalContext so agent sees validation warnings (52288a9)
• hooks: wrap main logic in try/except to guarantee exit 0 on partial edits (95ea24e)
• rcc: add Common Rationalizations section to 6 skills (Law 7 compliance) (6f2e9a4)
• rcc: add opus for planning/architecture in model guide (cb00637)
• rcc: correct rule-reviewer frontmatter fields (d0a78f7)
• rcc: migrating-plugins — add plugin discovery, selection, path resolution (ebf28f2)
• rcc: remove explicit hooks reference — auto-loaded by convention (8385b95)
• rcc: remove invalid context:fork from agents, drop context from allowed fields (12ce674)
• rcc: rewrite initializing-projects — fix blueprint, routing, validation (f8a2b4d)
• rcc: update category range 1-10 to 1-11 in analyzing-agent-systems (882d5bf)
• rcc: update model guide — sonnet for implementation, haiku for lookup (b4ff137)
• remove explicit skills array from plugin.json for auto-discovery (f8e71a2)
• resolve pylint issues and simplify README (16caec2)
• update Law 3 to enforce rule usage and creation (a0dcb70)
• v8.0.3 - require full detail presentation before user confirmation (b7f7af9)
• 修正 settings.json 驗證器並更新 hook schema 至最新規格 (413bbde)
• 刪除無意義的測試劇場 (8039a87)
• 用詞修正 - 智能改為智慧 (bf30f31)
• 統一 agent system 設定入口為 migrating-agent-systems (1ea3ff0)