cm-ux-master
starsforks
How to install cm-ux-master?
tody-agent/codymaster/cm-ux-master
$npx -y skills add tody-agent/codymaster --skill cm-ux-masterWorking memory protocol — maintains context across sessions via CONTINUITY.md. Inspired by Loki Mode. Read at turn start, update at turn end. Captures mistakes and learnings to prevent repeating errors.
SKILL.md
| 1 | # Continuity — Working Memory Protocol |
| 2 | |
| 3 | > **Context persistence across sessions. Mistakes captured. Learnings applied.** |
| 4 | > Inspired by Loki Mode's CONTINUITY.md protocol (Autonomi). |
| 5 | |
| 6 | ## When to Use |
| 7 | |
| 8 | **ALWAYS** — This is a background protocol, not an explicit invocation. |
| 9 | |
| 10 | - **Start of every session:** Read `.cm/CONTINUITY.md` to orient yourself |
| 11 | - **End of every session:** Update `.cm/CONTINUITY.md` with progress |
| 12 | - **On error:** Record in Mistakes & Learnings section |
| 13 | - **On key decision:** Record in Key Decisions section |
| 14 | |
| 15 | ## Setup |
| 16 | |
| 17 | ```bash |
| 18 | # Initialize working memory for current project |
| 19 | cm continuity init |
| 20 | |
| 21 | # Check current state |
| 22 | cm continuity status |
| 23 | |
| 24 | # View captured learnings |
| 25 | cm continuity learnings |
| 26 | ``` |
| 27 | |
| 28 | ## The Protocol |
| 29 | |
| 30 | ### AT THE START OF EVERY SESSION: |
| 31 | |
| 32 | ``` |
| 33 | 1. Read .cm/CONTINUITY.md to understand current state |
| 34 | 2. Read "Mistakes & Learnings" to avoid past errors |
| 35 | 3. Check "Next Actions" to determine what to do |
| 36 | 4. Reference Active Goal throughout your work |
| 37 | ``` |
| 38 | |
| 39 | ### DURING WORK: |
| 40 | |
| 41 | ``` |
| 42 | PRE-ACT ATTENTION CHECK (before every significant action): |
| 43 | - Re-read Active Goal |
| 44 | - Ask: "Does my planned action serve this goal?" |
| 45 | - Ask: "Am I solving the original problem, not a tangent?" |
| 46 | - If DRIFT detected → log it → return to goal |
| 47 | ``` |
| 48 | |
| 49 | ### AT THE END OF EVERY SESSION: |
| 50 | |
| 51 | ``` |
| 52 | 1. Update "Just Completed" with accomplishments |
| 53 | 2. Update "Next Actions" with remaining work |
| 54 | 3. Record any new "Mistakes & Learnings" |
| 55 | 4. Record any "Key Decisions" made |
| 56 | 5. Update "Files Modified" list |
| 57 | 6. Set currentPhase and timestamp |
| 58 | ``` |
| 59 | |
| 60 | ### ON ERROR (Self-Correction Loop): |
| 61 | |
| 62 | ``` |
| 63 | ON_ERROR: |
| 64 | 1. Capture error details (stack trace, context) |
| 65 | 2. Analyze root cause (not just symptoms) |
| 66 | 3. Write learning to CONTINUITY.md "Mistakes & Learnings" |
| 67 | 4. Update approach based on learning |
| 68 | 5. Retry with corrected approach |
| 69 | 6. Max 3 retries per error pattern before ESCALATE |
| 70 | ``` |
| 71 | |
| 72 | ## CONTINUITY.md Template |
| 73 | |
| 74 | ```markdown |
| 75 | # CodyMaster Working Memory |
| 76 | Last Updated: [ISO timestamp] |
| 77 | Current Phase: [planning|executing|testing|deploying|reviewing] |
| 78 | Current Iteration: [number] |
| 79 | Project: [project name] |
| 80 | |
| 81 | ## Active Goal |
| 82 | [What we're currently trying to accomplish — 1-2 sentences max] |
| 83 | |
| 84 | ## Current Task |
| 85 | - ID: [task-id from dashboard] |
| 86 | - Title: [task title] |
| 87 | - Status: [in-progress|blocked|reviewing] |
| 88 | - Skill: [cm-skill being used] |
| 89 | - Started: [timestamp] |
| 90 | |
| 91 | ## Just Completed |
| 92 | - [Most recent accomplishment with file:line references] |
| 93 | - [Previous accomplishment] |
| 94 | - [etc — last 5 items] |
| 95 | |
| 96 | ## Next Actions (Priority Order) |
| 97 | 1. [Immediate next step] |
| 98 | 2. [Following step] |
| 99 | 3. [etc] |
| 100 | |
| 101 | ## Active Blockers |
| 102 | - [Any current blockers or waiting items] |
| 103 | |
| 104 | ## Key Decisions This Session |
| 105 | - [Decision]: [Rationale] — [timestamp] |
| 106 | |
| 107 | ## Mistakes & Learnings |
| 108 | |
| 109 | ### Pattern: Error → Learning → Prevention |
| 110 | - **What Failed:** [Specific error that occurred] |
| 111 | - **Why It Failed:** [Root cause analysis] |
| 112 | - **How to Prevent:** [Concrete action to avoid this in future] |
| 113 | - **Timestamp:** [When learned] |
| 114 | - **Agent:** [Which agent] |
| 115 | - **Task:** [Which task ID] |
| 116 | |
| 117 | ## Working Context |
| 118 | [Critical information for current work — API keys paths, |
| 119 | architecture decisions, patterns being followed] |
| 120 | |
| 121 | ## Files Currently Being Modified |
| 122 | - [file path]: [what we're changing] |
| 123 | ``` |
| 124 | |
| 125 | ## Memory Hierarchy |
| 126 | |
| 127 | The memory systems complement each other: |
| 128 | |
| 129 | ``` |
| 130 | 1. CONTINUITY.md = Working memory (current session state) |
| 131 | 2. learnings.json = Extracted error patterns (persists across sessions) |
| 132 | 3. decisions.json = Architecture decisions (persists across projects) |
| 133 | 4. tasks.md/cm-tasks.json = Task queue + RARV logs (Mode D integration) |
| 134 | ``` |
| 135 | |
| 136 | **CONTINUITY.md is the PRIMARY source of truth for "what am I doing right now?"** |
| 137 | |
| 138 | ## Integration |
| 139 | |
| 140 | | Skill | How it integrates | |
| 141 | |-------|-------------------| |
| 142 | | `cm-execution` | RARV Mode D reads CONTINUITY.md in REASON phase | |
| 143 | | `cm-planning` | Sets Active Goal and Next Actions | |
| 144 | | `cm-debugging` | Records errors in Mistakes & Learnings | |
| 145 | | `cm-quality-gate` | VERIFY phase updates CONTINUITY.md | |
| 146 | | `cm-code-review` | Records review feedback as learnings | |
| 147 | |
| 148 | ## Rules |
| 149 | |
| 150 | ``` |
| 151 | ✅ DO: |
| 152 | - Read CONTINUITY.md at session start (ALWAYS) |
| 153 | - Update CONTINUITY.md at session end (ALWAYS) |
| 154 | - Record EVERY error in Mistakes & Learnings |
| 155 | - Keep "Just Completed" to last 5 items |
| 156 | - Be specific: "Fixed auth bug in login.ts:42" not "Fixed stuff" |
| 157 | |
| 158 | ❌ DON'T: |
| 159 | - Skip reading CONTINUITY.md ("I remember what I was doing") |
| 160 | - Write vague learnings: "It didn't work" → WHY didn't it work? |
| 161 | - Ignore past learnings when they're relevant |
| 162 | - Let CONTINUITY.md grow beyond ~500 words (rotate old entries) |
| 163 | - Delete Mistakes & Learnings (archive to learnings.json instead) |
| 164 | ``` |
| 165 | |
| 166 | ## The Bottom Line |
| 167 | |
| 168 | **Your memory is your superpower. Without it, you repeat every mistake forever.** |
People who liked this also viewed
6 picksType
- web-design-guidelinesReview UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check accessibility", "audit design", "review UX", or "check my site against best practices".SkillsJul 2026
- lark-slides飞书幻灯片：创建和编辑幻灯片。创建演示文稿、读取幻灯片内容、管理幻灯片页面（创建、删除、读取、局部替换）。当用户需要创建或编辑幻灯片、读取或修改单个页面时使用。当用户给出 doubao.com 的 /slides/ URL/token 时，也应直接使用本 skill，不要因为域名不是飞书而回退到 WebFetch；路由依据是 URL 路径模式和 token，而不是域名。不负责：云文档内容编辑（走 SkillsJul 2026
- design-taste-frontendTaste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop SkillsJul 2026
- ui-ux-pro-maxUI/UX design intelligence for web and mobile. Searchable local database with 84 styles, 192 color palettes, 74 font pairings, 192 product types, 98 UX guidelines, 104 icon entries, 16 GSAP motion presSkillsJul 2026
- caveman-helpQuick-reference card for all caveman modes, skills, and commands. One-shot display, not a persistent mode. Trigger: /caveman-help, "caveman help", "what caveman commands", "how do I use caveman".SkillsJul 2026
- codebase-designShared vocabulary for designing deep modules. Use when the user wants to design or improve a module's interface, find deepening opportunities, decide where a seam goes, make code more testable or AI-nSkillsJul 2026
