[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MFND8Q1VEEBMGPRZ5AM12E":3,"skill-same:01M6MFND8Q1VEEBMGPRZ5AM12E":124},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":45,"linkOnly":46,"markdownContent":47,"markdownBody":48,"frontMatter":49,"refName":50,"commitSha":51,"skillDirectory":52,"forksCount":53,"repositoryHtmlUrl":54,"repositoryHomepageUrl":55,"ownerHtmlUrl":56,"ownerBlogUrl":57,"ownerDescription":58,"updatedAt":59,"install":60,"resources":73},"01M6MFND8Q1VEEBMGPRZ5AM12E","vibe-portrait",null,"Developer personality portrait generator. Supports subcommands: (1) 'generate my portrait' \u002F 'analyze my personality' — full analysis; (2) 'update my portrait' — incremental update since last analysis; (3) 'install persona from \u003Curl>' — install community persona from GitHub; (4) 'list personas' \u002F '我安装了哪些人格' — show installed; (5) 'remove persona \u003Cid>' — uninstall; (6) 'think like \u003Cname>' \u002F '像\u003Cname>一样思考' — activate a persona. Also triggers on 'vibe-portrait', 'what kind of developer am I'.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1575,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000030","开发工程师","PROFESSION","适用职业",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44],"vibe portrait","vibe","portrait","personas","think like","kind","generator","URL","name","id","triggers","full","github","activate","show","remove","list","analyze","developer","update","generate","2026-11-10 10:12:30",false,"---\nname: \"vibe-portrait\"\ndescription: \"Developer personality portrait generator. Supports subcommands: (1) 'generate my portrait' \u002F 'analyze my personality' — full analysis; (2) 'update my portrait' — incremental update since last analysis; (3) 'install persona from \u003Curl>' — install community persona from GitHub; (4) 'list personas' \u002F '我安装了哪些人格' — show installed; (5) 'remove persona \u003Cid>' — uninstall; (6) 'think like \u003Cname>' \u002F '像\u003Cname>一样思考' — activate a persona. Also triggers on 'vibe-portrait', 'what kind of developer am I'.\"\n---\n\n# Vibe Portrait\n\nDeveloper personality portrait generator with subcommands.\n\n## Command routing\n\nDetermine which subcommand the user wants based on their message:\n\n| Trigger | Action |\n|---------|--------|\n| \"generate my portrait\" \u002F \"analyze my personality\" \u002F just \"\u002Fvibe-portrait\" | → **Generate** (full analysis, Steps 0-8) |\n| \"update my portrait\" \u002F \"update portrait\" \u002F \"更新我的画像\" | → **Update** (incremental, see Update section) |\n| \"install persona from \\\u003Curl\\>\" \u002F \"安装人格 \\\u003Curl\\>\" | → **Install** (see Persona Management) |\n| \"list personas\" \u002F \"我安装了哪些人格\" | → **List** (see Persona Management) |\n| \"remove persona \\\u003Cid\\>\" \u002F \"删除人格 \\\u003Cid\\>\" | → **Remove** (see Persona Management) |\n| \"think like \\\u003Cname\\>\" \u002F \"像\\\u003Cname\\>一样思考\" | → **Activate** (load the persona skill, not handled here — Claude auto-activates from the installed skill) |\n\nIf ambiguous, ask the user which action they want.\n\n---\n\n## Generate (full analysis)\n\n## Step 0: Ask analysis mode\n\nBefore reading any data, ask the user which analysis mode they prefer:\n\n> **How thorough should the analysis be?**\n>\n> 1. **⚡ Quick (cost-efficient)** — Sample ~200 messages. Fast, low token cost, good enough for most portraits.\n> 2. **? Full (comprehensive)** — Read ALL messages. Higher token cost, but captures every nuance and evolution of your personality. Best for users with long history who want maximum accuracy.\n\nIf the user doesn't respond or says \"just do it\" \u002F \"默认\" \u002F \"whatever\", default to **Quick mode**.\n\nIn **Quick mode**, follow the sampling strategy described in Step 1c.\nIn **Full mode**, read all lines from `history.jsonl` (and any imported files). Skip the sampling strategy entirely. If the file exceeds 3000 lines, still read in batches (e.g., 500 lines at a time) to avoid tool errors, but process every line.\n\n## Step 1: Locate and read conversation data\n\n### 1a: Local history\n\n**Always read from ALL available sources regardless of which terminal the user is running.** A developer's personality is shaped by all their AI interactions, not just one tool.\n\nCheck and read from every source that exists:\n\n| Source | Path | Message field |\n|--------|------|---------------|\n| Claude Code history | `~\u002F.claude\u002Fhistory.jsonl` | `display` |\n| Claude Code projects | `~\u002F.claude\u002Fprojects\u002F**\u002F*.jsonl` | `display` |\n| Codex history | `~\u002F.codex\u002Fhistory.jsonl` | `text` (also has `ts` unix timestamp) |\n| Codex sessions | `~\u002F.codex\u002Fsessions\u002FYYYY\u002FMM\u002FDD\u002Frollout-*.jsonl` | filter `type=response_item` + `payload.role=user` → iterate `payload.content[]`, take items where `text` does NOT start with `\u003C` (skip system-injected `\u003Cenvironment_context>`, `\u003Cpermissions>` etc.) |\n\nMerge all messages from all sources into one pool before sampling. Use `ts` timestamps (Codex history) or `timestamp` (Codex sessions) for chronological sorting when available.\n\n### 1b: Multi-machine sync via Git repo\n\nVibe Portrait uses a **private GitHub repo** to sync persona data across machines. This is the recommended approach for users with multiple development machines.\n\n**First-time setup (if no portrait repo exists yet):**\n\nAfter analysis is complete (Step 7), the skill will offer to create a private repo. See Step 7 for details.\n\n**If the user already has a portrait repo:**\n\nAsk the user for their portrait repo URL, or check if `~\u002F.vibe-portrait-repo` exists (a file containing the repo URL, written during first-time setup).\n\n```bash\n# Check for existing repo config\ncat ~\u002F.vibe-portrait-repo 2>\u002Fdev\u002Fnull\n```\n\nIf found, clone or pull the repo into a temp directory:\n```bash\n# Clone if not already local\nREPO_URL=$(cat ~\u002F.vibe-portrait-repo)\nREPO_DIR=~\u002F.cache\u002Fvibe-portrait-sync\ngit clone \"$REPO_URL\" \"$REPO_DIR\" 2>\u002Fdev\u002Fnull || (cd \"$REPO_DIR\" && git pull)\n```\n\nThen load existing analysis files from `$REPO_DIR\u002Fanalysis\u002F*.json`. When merging with the current machine's analysis:\n- Each machine's analysis is stored as a separate JSON file: `analysis\u002F{hostname}-{date}.json`\n- To merge: average the numeric scores across all machine analyses, weighted by message count\n- Union all quotes (deduplicate by text)\n- Union all detected domains and tools\n- The persona skill (`me\u002FSKILL.md`) is always regenerated from the merged result\n\n**If the user provides a `.jsonl` file directly (fallback):**\n\nRead it as a supplementary source. Merge messages from all sources before sampling:\n- Deduplicate messages with identical `display` content\n- If timestamps are available, sort merged messages chronologically\n- Treat all sources equally\n\n### 1c: Read strategy\n\nUse a bash command to count total lines first:\n```bash\nwc -l ~\u002F.claude\u002Fhistory.jsonl\n```\n\n**Quick mode (default):** Sample strategically:\n- Read the **first 50 lines** (early personality)\n- Read the **last 100 lines** (current personality)\n- Read **50 random lines from the middle** (evolution)\n- Total: ≤200 messages per source file\n\n**Full mode:** Read all lines. For large files (>500 lines), read in batches of 500 lines using offset\u002Flimit to avoid tool errors. Process every message.\n\n**Parse each line as JSON and extract the `display` field.** Skip lines where `display` is empty, null, or a slash command (starts with `\u002F`).\n\n**Minimum threshold:** If fewer than 20 non-empty messages are found across all sources, inform the user that there isn't enough data for a meaningful portrait. Suggest they try again after more conversations.\n\n## Step 2: Analyze across six dimensions\n\nRead the analysis framework for detailed signal definitions:\n→ `references\u002Fanalysis-framework.md`\n\nAnalyze across six dimensions: Communication Style, Technical Breadth, Technical Depth, Decision Patterns, Collaboration Style, Work Rhythm. See the reference file for detailed signal definitions.\n\n**Language detection (for i18n)**: if >50% of sampled messages are in Chinese, set `meta.lang = \"zh\"`. Otherwise `\"en\"`. This determines the portrait page language.\n\n**For each dimension, produce:** a score (0-100), a one-line summary, and 2-3 evidence quotes (keep quotes short, anonymize sensitive content).\n\n## Step 3: Compute personality type, rating, and famous match\n\n### 3.1 MBTI Mapping\nRead: → `references\u002Fmbti-mapping.md`\n\nMap observed signals to four axes:\n- **E\u002FI**: delegation breadth vs self-reliance\n- **S\u002FN**: implementation focus vs architecture focus\n- **T\u002FF**: logic-driven vs people-aware\n- **J\u002FP**: structured planning vs exploratory iteration\n\nEach axis: 0-100 score. Letter determined by which side of 50.\n\n### 3.2 Developer Rating\nRead: → `references\u002Frating-rubric.md`\n\nRate on a six-tier scale (provide BOTH `label` for Chinese and `labelEn` for English):\n\n| Tier | label (zh) | labelEn (en) | Emoji |\n|------|-----------|--------------|-------|\n| S+ | 夯爆了 | Legendary | ? |\n| S | 夯 | Elite | ? |\n| A | 人上人 | Above Average | ⭐ |\n| B | NPC | NPC | ? |\n| C | 拉 | Below Average | ? |\n| D | 拉完了 | Inactive | ? |\n\nAlways include strength highlights AND growth areas, even for top-tier ratings.\n\n### 3.3 Famous Person Match\nRead: → `references\u002Ffamous-matching.md`\n\nMatch the user across 3 independent dimensions (Technical Spirit \u002F Strategic Mind \u002F Communication Soul), each picking a different historical or contemporary figure. No predefined list — the AI uses its own knowledge. See reference file for rules and anti-bias constraints.\n\n## Step 4: Generate the portrait HTML\n\nRead the HTML template:\n→ `templates\u002Fportrait.html`\n\nThe template contains a `PORTRAIT_DATA` JavaScript object near the top of the `\u003Cbody>`. Replace the ENTIRE `PORTRAIT_DATA` object with the computed data. The structure is:\n\n```javascript\nconst PORTRAIT_DATA = {\n  meta: {\n    username: \"...\",         \u002F\u002F infer from system username or ask\n    generatedAt: \"...\",      \u002F\u002F today's date\n    messageCount: 0,         \u002F\u002F actual count of sampled messages\n    dateRange: \"...\",        \u002F\u002F earliest to latest message date\n    confidence: \"...\",       \u002F\u002F \"Early Sketch\" \u002F \"Clear Portrait\" \u002F \"Deep Portrait\"\n    lang: \"...\"              \u002F\u002F \"zh\" if user's messages are predominantly Chinese, otherwise \"en\"\n  },\n  personality: {\n    mbtiType: \"XXXX\",\n    axes: {\n      EI: { score: 0, letter: \"X\", label: \"...\" },\n      SN: { score: 0, letter: \"X\", label: \"...\" },\n      TF: { score: 0, letter: \"X\", label: \"...\" },\n      JP: { score: 0, letter: \"X\", label: \"...\" }\n    },\n    summary: \"...\"\n  },\n  radar: {\n    technicalDepth: 0,\n    technicalBreadth: 0,\n    communication: 0,\n    decisionSpeed: 0,\n    collaboration: 0,\n    creativity: 0\n  },\n  rating: {\n    tier: \"...\",\n    label: \"...\",             \u002F\u002F Chinese label: 夯爆了\u002F夯\u002F人上人\u002FNPC\u002F拉\u002F拉完了\n    labelEn: \"...\",           \u002F\u002F English label: Legendary\u002FElite\u002FAbove Average\u002FNPC\u002FBelow Average\u002FInactive\n    emoji: \"...\",\n    color: \"...\",\n    reason: \"...\",\n    strengthHighlights: [],\n    growthAreas: []\n  },\n  famousMatch: {\n    technical: {\n      name: \"...\", emoji: \"...\",\n      dimension: \"Technical Spirit\", dimensionZh: \"技术灵魂\",\n      reason: \"...\", sharedTraits: []\n    },\n    strategic: {\n      name: \"...\", emoji: \"...\",\n      dimension: \"Strategic Mind\", dimensionZh: \"思维内核\",\n      reason: \"...\", sharedTraits: []\n    },\n    communication: {\n      name: \"...\", emoji: \"...\",\n      dimension: \"Communication Soul\", dimensionZh: \"表达人格\",\n      reason: \"...\", sharedTraits: []\n    }\n  },\n  communication: {\n    languages: {},\n    directnessScore: 0,\n    avgMessageLength: 0,\n    topKeywords: [],\n    questionRatio: 0\n  },\n  technical: {\n    domains: {},\n    topTools: []\n  },\n  workRhythm: {\n    hourlyActivity: [\u002F* 24 integers, one per hour *\u002F],\n    sessionPattern: \"...\",\n    avgSessionLength: \"...\",\n    projectCount: 0\n  },\n  quotes: [\n    { text: \"...\", context: \"...\" }\n    \u002F\u002F 3-5 representative quotes\n  ]\n};\n```\n\n**IMPORTANT:**\n- Replace ONLY the `PORTRAIT_DATA` object. Do not modify any other part of the template.\n- All values must be valid JavaScript (strings quoted, arrays bracketed, no trailing commas).\n- The rest of the template's JavaScript will automatically render everything from this object.\n- Keep quote texts under 100 characters. Anonymize any sensitive data (API keys, tokens, passwords, personal URLs).\n\n## Step 5: Generate persona skill\n\nRead the persona skill template:\n→ `references\u002Fpersona-skill-template.md`\n\nGenerate a **multi-file** persona skill. Read the template for the full structure:\n→ `references\u002Fpersona-skill-template.md`\n\nCreate the following files under `~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002F`:\n\n```\nme\u002F\n├── SKILL.md                        # Entry point (~80 lines)\n├── portrait-meta.json              # Timestamps for incremental updates\n└── references\u002F\n    ├── thinking-patterns.md        # From: technical depth + decision patterns\n    ├── decision-framework.md       # From: decision patterns + collaboration style\n    ├── communication-style.md      # From: communication style analysis\n    ├── engineering-philosophy.md   # From: technical depth + breadth + anti-patterns\n    └── mindset-markers.md          # Abstracted attitudes (NO raw quotes)\n```\n\n**`portrait-meta.json` is critical** — it stores `lastMessageIndex` (line count of history.jsonl at analysis time) and `updatedAt` timestamp. These enable the **update** subcommand to do incremental analysis later.\n\nCreate directories if they don't exist. `me\u002F` is reserved for the user's own persona.\n\n## Step 6: Write output files\n\nWrite the following files to the current working directory:\n\n**1. HTML Portrait:**\n```\nvibe-portrait-YYYY-MM-DD.html\n```\n\n**2. Portrait Image (auto-export):**\n\nAfter writing the HTML, attempt to generate a PNG screenshot of the portrait. Try these methods in order:\n\n- **If Playwright MCP is available** (check for `playwright_navigate` and `playwright_screenshot` tools):\n  1. Navigate to the local HTML file: `file:\u002F\u002F\u002Fabsolute\u002Fpath\u002Fto\u002Fvibe-portrait-YYYY-MM-DD.html`\n  2. Wait for charts to render (2-3 seconds)\n  3. Take a full-page screenshot, save as `vibe-portrait-YYYY-MM-DD.png`\n\n- **If Playwright is not available**, try bash:\n  ```bash\n  # Try with npx playwright if installed\n  npx --yes playwright screenshot --full-page \"file:\u002F\u002F$(pwd)\u002Fvibe-portrait-YYYY-MM-DD.html\" \"vibe-portrait-YYYY-MM-DD.png\" 2>\u002Fdev\u002Fnull\n  ```\n\n- **If neither works**, skip silently. Tell the user they can use the \"Export as Image\" button on the HTML page instead.\n\n**3. Analysis JSON** (for sync\u002Fmerge):\n```\nvibe-portrait-analysis-YYYY-MM-DD.json\n```\n\nContents:\n```json\n{\n  \"version\": \"1.0\",\n  \"machine\": \"\u003Chostname from `hostname` command>\",\n  \"exportedAt\": \"YYYY-MM-DD\",\n  \"sourceFiles\": [\"~\u002F.claude\u002Fhistory.jsonl\"],\n  \"messageCount\": 200,\n  \"portraitData\": { \u002F* the full PORTRAIT_DATA object *\u002F }\n}\n```\n\n**3. Persona skill** (also written in Step 5):\n```\n~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002FSKILL.md\n```\n\n## Step 7: Sync to portrait repo\n\nAfter writing all output files, ask the user:\n\n> **Sync to your portrait repo?**\n>\n> 1. **Yes — create a new repo** (first time)\n> 2. **Yes — push to existing repo**\n> 3. **No — keep everything local**\n\n### Option 1: Create new portrait repo\n\n1. Check `gh auth status`. If not authenticated, tell user to run `! gh auth login`.\n2. Copy `repo-template\u002F` contents into a temp directory.\n3. Copy the generated analysis JSON, HTML portrait, and persona skill into the temp directory under `analysis\u002F`, `portraits\u002F`, and `me\u002F` respectively.\n4. **Fill in the README.md** — the template contains `{{PLACEHOLDER}}` markers. Read `repo-template\u002FREADME.md` (see the file for the full list of placeholders) and replace them all with the user's analysis data. Key formatting rules:\n   - Badge lists → inline code: `` `item1` `item2` ``\n   - Domain bars → Unicode blocks: `**ML\u002FAI** ██████████████░░░░░░ 30%` (20 chars wide, `█` filled, `░` empty)\n   - Quotes → blockquotes: `> \"text\" — *context*`\n   - Language breakdown → `\"Chinese 65% · English 30% · Mixed 5%\"`\n5. `git init && git add -A && git commit -m \"Initial portrait\"`\n6. `gh repo create my-vibe-portrait --private --source=. --push`\n7. Save repo URL: `gh repo view --json url -q '.url' > ~\u002F.vibe-portrait-repo`\n\n### Option 2: Push to existing repo\n\n1. Read repo URL from `~\u002F.vibe-portrait-repo`.\n2. Clone (or pull if already cached at `~\u002F.cache\u002Fvibe-portrait-sync`).\n3. Copy the latest analysis JSON, HTML, and persona skill into the repo directory.\n4. Regenerate `README.md` using the same placeholder-filling process as Option 1.\n5. `git add -A && git commit -m \"Update portrait from $(hostname)\" && git push`\n\n### Option 3: Keep local\n\nDo nothing. Files are already written locally.\n\n## Step 8: Present results\n\nTell the user:\n1. Where the HTML portrait was saved (open in browser)\n2. Where the persona skill was installed (`~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002F`)\n3. Whether the portrait repo was synced (and the repo URL if applicable)\n4. A brief summary of key findings (MBTI type, rating, famous match)\n5. How to use the persona: **\"think like me\"** or **\"像我一样思考\"**\n6. How to sync from another machine: install vibe-portrait there, run it, choose \"push to existing repo\"\n\n## Confidence levels\n\nBased on total non-empty messages found:\n- **\u003C 20**: Do not generate. Tell user to accumulate more history.\n- **20-50**: Generate with `confidence: \"Early Sketch\"`. Warn that results are approximate.\n- **50-200**: Generate with `confidence: \"Clear Portrait\"`.\n- **200+**: Generate with `confidence: \"Deep Portrait\"`.\n\n## Output contract\n\nThis skill produces THREE outputs:\n\n### 1. HTML Portrait (`vibe-portrait-YYYY-MM-DD.html`)\n- Single self-contained file (only external dependencies: Tailwind CDN + Chart.js CDN)\n- Opens correctly in any modern browser\n- Contains all analysis data in the `PORTRAIT_DATA` JavaScript object\n- Includes actual quotes (anonymized for sensitive content)\n\n### 2. Persona Skill (`~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002FSKILL.md`)\n- Valid Claude Code \u002F Codex skill file with proper YAML frontmatter\n- Under 200 lines, dense and actionable\n- Activates on \"think like me\" \u002F \"像我一样思考\" and similar phrases\n- Captures thinking patterns, decision framework, communication style, engineering philosophy\n\n### 3. Analysis JSON (`vibe-portrait-analysis-YYYY-MM-DD.json`)\n- Portable data file for cross-machine syncing\n- Contains full PORTRAIT_DATA plus machine hostname and export metadata\n- Synced to portrait repo (if configured) or kept local\n\n### 4. Portrait Repo (optional, private GitHub repo)\n- Created via `gh repo create my-vibe-portrait --private`\n- Stores persona skill, analysis JSONs from all machines, and latest portrait\n- Repo URL cached in `~\u002F.vibe-portrait-repo` for future syncs\n- Template provided in `repo-template\u002F` within this skill\n\n## Privacy rules (CRITICAL — read before generating ANY output)\n\nAll outputs of VibePortrait (HTML, persona skill, analysis JSON, repo README) are potentially shareable. Apply these rules to **every file you write**:\n\n### What MUST be redacted\n- **API keys, tokens, passwords, secrets** → replace with `[REDACTED]`\n- **URLs containing credentials** (tokens in query strings, auth headers) → redact the credential portion\n- **File paths exposing personal info** (`\u002Fhome\u002Fusername\u002F`, `\u002FUsers\u002Frealname\u002F`) → replace with `~\u002F` or `\u003Chome>\u002F`\n- **Email addresses, phone numbers, IPs** → redact unless the user explicitly consents\n- **Private repo URLs, internal hostnames** → redact\n\n### What MUST NOT appear in persona skills\n- **No verbatim chat messages.** Persona reference files must contain only abstracted personality descriptions, never raw conversation text.\n- **No project names, company names, or colleague names** from the user's conversations\n- **No financial data** (account balances, trading strategies, PnL numbers)\n- **No credentials or deployment details** (server IPs, Docker configs, database names)\n\n### What is OK\n- Abstracted personality traits, thinking patterns, communication style descriptions\n- Generic domain labels (\"ML\u002FAI\", \"Web\", \"Systems\u002FInfra\")\n- Tool\u002Fframework names (public knowledge: \"PyTorch\", \"Docker\", \"React\")\n- MBTI scores, radar scores, rating tier (all computed, not raw data)\n\n### Before pushing to git repo\nScan all files being committed for the redaction patterns above. If any slip through, fix before pushing.\n\n## Anti-patterns\n\n- **Do not fabricate data.** If you can't determine a signal, use a neutral score (50).\n- **Do not read files beyond conversation history.** No source code, private documents, SSH keys.\n- **Do not skip the template.** Always use `templates\u002Fportrait.html` as the base.\n- **Do not read all history lines in Quick mode.** Use the sampling strategy.\n\n## Reference files\n\n`references\u002Fanalysis-framework.md` · `references\u002Fmbti-mapping.md` · `references\u002Ffamous-matching.md` · `references\u002Frating-rubric.md` · `references\u002Fpersona-skill-template.md` · `templates\u002Fportrait.html` · `repo-template\u002F`\n\n## Update (incremental analysis)\n\nWhen the user says \"update my portrait\" \u002F \"更新我的画像\":\n\n1. Read `~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002Fportrait-meta.json`.\n   - If it doesn't exist, tell the user to run a full generate first.\n2. Get `lastMessageIndex` from the meta file.\n3. Count current lines in `~\u002F.claude\u002Fhistory.jsonl`.\n4. If no new messages since last analysis, tell the user \"Portrait is up to date.\"\n5. Otherwise, read only the **new messages** (from line `lastMessageIndex+1` to end).\n6. Analyze the new messages across the same 6 dimensions.\n7. **Merge** with existing analysis: weighted average of scores by message count (old weight = `totalMessagesAnalyzed`, new weight = new message count).\n8. Regenerate: HTML portrait, all persona reference files, portrait-meta.json (update `updatedAt`, `lastMessageIndex`, `totalMessagesAnalyzed`).\n9. If portrait repo exists, push the update.\n\nThis is much cheaper than a full re-analysis — typically only processes dozens of new messages.\n\n---\n\n## Persona Management\n\n### Install from GitHub\n\nTrigger: \"install persona from \\\u003Curl\\>\"\n\n```bash\nTEMP_DIR=$(mktemp -d)\ngit clone --depth 1 \"\u003Curl>\" \"$TEMP_DIR\"\nPERSON_ID=$(grep -m1 'personaId' \"$TEMP_DIR\u002Fme\u002Fportrait-meta.json\" | sed 's\u002F.*: *\"\\(.*\\)\".*\u002F\\1\u002F' )\n# Fallback: infer from GitHub username\nTARGET=~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002F$PERSON_ID\ncp -R \"$TEMP_DIR\u002Fme\" \"$TARGET\"\nrm -rf \"$TEMP_DIR\"\n```\n\nConfirm: where installed + how to invoke (`think like {person-id}`).\n\n### List\n\nTrigger: \"list personas\" \u002F \"我安装了哪些人格\"\n\n```bash\nls ~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002F\n```\n\nFor each directory, read `portrait-meta.json` and show: name, MBTI type, rating, last updated.\n\n### Remove\n\nTrigger: \"remove persona \\\u003Cid\\>\"\n\n```bash\nrm -rf ~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002F\u003Cid>\n```\n\nConfirm before deleting. Refuse to delete `me\u002F` — tell user to run generate to overwrite instead.\n\n### Directory layout\n\n```\n~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002F\n├── me\u002F                        ← yours (multi-file, auto-generated)\n│   ├── SKILL.md\n│   ├── portrait-meta.json\n│   └── references\u002F\n│       ├── thinking-patterns.md\n│       ├── decision-framework.md\n│       ├── communication-style.md\n│       ├── engineering-philosophy.md\n│       └── mindset-markers.md\n├── jane-doe\u002F                  ← installed from GitHub\n│   ├── SKILL.md\n│   ├── portrait-meta.json\n│   └── references\u002F...\n└── zhuge-liang\u002F               ← installed from community\n    └── ...\n","# Vibe Portrait\n\nDeveloper personality portrait generator with subcommands.\n\n## Command routing\n\nDetermine which subcommand the user wants based on their message:\n\n| Trigger | Action |\n|---------|--------|\n| \"generate my portrait\" \u002F \"analyze my personality\" \u002F just \"\u002Fvibe-portrait\" | → **Generate** (full analysis, Steps 0-8) |\n| \"update my portrait\" \u002F \"update portrait\" \u002F \"更新我的画像\" | → **Update** (incremental, see Update section) |\n| \"install persona from \\\u003Curl\\>\" \u002F \"安装人格 \\\u003Curl\\>\" | → **Install** (see Persona Management) |\n| \"list personas\" \u002F \"我安装了哪些人格\" | → **List** (see Persona Management) |\n| \"remove persona \\\u003Cid\\>\" \u002F \"删除人格 \\\u003Cid\\>\" | → **Remove** (see Persona Management) |\n| \"think like \\\u003Cname\\>\" \u002F \"像\\\u003Cname\\>一样思考\" | → **Activate** (load the persona skill, not handled here — Claude auto-activates from the installed skill) |\n\nIf ambiguous, ask the user which action they want.\n\n---\n\n## Generate (full analysis)\n\n## Step 0: Ask analysis mode\n\nBefore reading any data, ask the user which analysis mode they prefer:\n\n> **How thorough should the analysis be?**\n>\n> 1. **⚡ Quick (cost-efficient)** — Sample ~200 messages. Fast, low token cost, good enough for most portraits.\n> 2. **? Full (comprehensive)** — Read ALL messages. Higher token cost, but captures every nuance and evolution of your personality. Best for users with long history who want maximum accuracy.\n\nIf the user doesn't respond or says \"just do it\" \u002F \"默认\" \u002F \"whatever\", default to **Quick mode**.\n\nIn **Quick mode**, follow the sampling strategy described in Step 1c.\nIn **Full mode**, read all lines from `history.jsonl` (and any imported files). Skip the sampling strategy entirely. If the file exceeds 3000 lines, still read in batches (e.g., 500 lines at a time) to avoid tool errors, but process every line.\n\n## Step 1: Locate and read conversation data\n\n### 1a: Local history\n\n**Always read from ALL available sources regardless of which terminal the user is running.** A developer's personality is shaped by all their AI interactions, not just one tool.\n\nCheck and read from every source that exists:\n\n| Source | Path | Message field |\n|--------|------|---------------|\n| Claude Code history | `~\u002F.claude\u002Fhistory.jsonl` | `display` |\n| Claude Code projects | `~\u002F.claude\u002Fprojects\u002F**\u002F*.jsonl` | `display` |\n| Codex history | `~\u002F.codex\u002Fhistory.jsonl` | `text` (also has `ts` unix timestamp) |\n| Codex sessions | `~\u002F.codex\u002Fsessions\u002FYYYY\u002FMM\u002FDD\u002Frollout-*.jsonl` | filter `type=response_item` + `payload.role=user` → iterate `payload.content[]`, take items where `text` does NOT start with `\u003C` (skip system-injected `\u003Cenvironment_context>`, `\u003Cpermissions>` etc.) |\n\nMerge all messages from all sources into one pool before sampling. Use `ts` timestamps (Codex history) or `timestamp` (Codex sessions) for chronological sorting when available.\n\n### 1b: Multi-machine sync via Git repo\n\nVibe Portrait uses a **private GitHub repo** to sync persona data across machines. This is the recommended approach for users with multiple development machines.\n\n**First-time setup (if no portrait repo exists yet):**\n\nAfter analysis is complete (Step 7), the skill will offer to create a private repo. See Step 7 for details.\n\n**If the user already has a portrait repo:**\n\nAsk the user for their portrait repo URL, or check if `~\u002F.vibe-portrait-repo` exists (a file containing the repo URL, written during first-time setup).\n\n```bash\n# Check for existing repo config\ncat ~\u002F.vibe-portrait-repo 2>\u002Fdev\u002Fnull\n```\n\nIf found, clone or pull the repo into a temp directory:\n```bash\n# Clone if not already local\nREPO_URL=$(cat ~\u002F.vibe-portrait-repo)\nREPO_DIR=~\u002F.cache\u002Fvibe-portrait-sync\ngit clone \"$REPO_URL\" \"$REPO_DIR\" 2>\u002Fdev\u002Fnull || (cd \"$REPO_DIR\" && git pull)\n```\n\nThen load existing analysis files from `$REPO_DIR\u002Fanalysis\u002F*.json`. When merging with the current machine's analysis:\n- Each machine's analysis is stored as a separate JSON file: `analysis\u002F{hostname}-{date}.json`\n- To merge: average the numeric scores across all machine analyses, weighted by message count\n- Union all quotes (deduplicate by text)\n- Union all detected domains and tools\n- The persona skill (`me\u002FSKILL.md`) is always regenerated from the merged result\n\n**If the user provides a `.jsonl` file directly (fallback):**\n\nRead it as a supplementary source. Merge messages from all sources before sampling:\n- Deduplicate messages with identical `display` content\n- If timestamps are available, sort merged messages chronologically\n- Treat all sources equally\n\n### 1c: Read strategy\n\nUse a bash command to count total lines first:\n```bash\nwc -l ~\u002F.claude\u002Fhistory.jsonl\n```\n\n**Quick mode (default):** Sample strategically:\n- Read the **first 50 lines** (early personality)\n- Read the **last 100 lines** (current personality)\n- Read **50 random lines from the middle** (evolution)\n- Total: ≤200 messages per source file\n\n**Full mode:** Read all lines. For large files (>500 lines), read in batches of 500 lines using offset\u002Flimit to avoid tool errors. Process every message.\n\n**Parse each line as JSON and extract the `display` field.** Skip lines where `display` is empty, null, or a slash command (starts with `\u002F`).\n\n**Minimum threshold:** If fewer than 20 non-empty messages are found across all sources, inform the user that there isn't enough data for a meaningful portrait. Suggest they try again after more conversations.\n\n## Step 2: Analyze across six dimensions\n\nRead the analysis framework for detailed signal definitions:\n→ `references\u002Fanalysis-framework.md`\n\nAnalyze across six dimensions: Communication Style, Technical Breadth, Technical Depth, Decision Patterns, Collaboration Style, Work Rhythm. See the reference file for detailed signal definitions.\n\n**Language detection (for i18n)**: if >50% of sampled messages are in Chinese, set `meta.lang = \"zh\"`. Otherwise `\"en\"`. This determines the portrait page language.\n\n**For each dimension, produce:** a score (0-100), a one-line summary, and 2-3 evidence quotes (keep quotes short, anonymize sensitive content).\n\n## Step 3: Compute personality type, rating, and famous match\n\n### 3.1 MBTI Mapping\nRead: → `references\u002Fmbti-mapping.md`\n\nMap observed signals to four axes:\n- **E\u002FI**: delegation breadth vs self-reliance\n- **S\u002FN**: implementation focus vs architecture focus\n- **T\u002FF**: logic-driven vs people-aware\n- **J\u002FP**: structured planning vs exploratory iteration\n\nEach axis: 0-100 score. Letter determined by which side of 50.\n\n### 3.2 Developer Rating\nRead: → `references\u002Frating-rubric.md`\n\nRate on a six-tier scale (provide BOTH `label` for Chinese and `labelEn` for English):\n\n| Tier | label (zh) | labelEn (en) | Emoji |\n|------|-----------|--------------|-------|\n| S+ | 夯爆了 | Legendary | ? |\n| S | 夯 | Elite | ? |\n| A | 人上人 | Above Average | ⭐ |\n| B | NPC | NPC | ? |\n| C | 拉 | Below Average | ? |\n| D | 拉完了 | Inactive | ? |\n\nAlways include strength highlights AND growth areas, even for top-tier ratings.\n\n### 3.3 Famous Person Match\nRead: → `references\u002Ffamous-matching.md`\n\nMatch the user across 3 independent dimensions (Technical Spirit \u002F Strategic Mind \u002F Communication Soul), each picking a different historical or contemporary figure. No predefined list — the AI uses its own knowledge. See reference file for rules and anti-bias constraints.\n\n## Step 4: Generate the portrait HTML\n\nRead the HTML template:\n→ `templates\u002Fportrait.html`\n\nThe template contains a `PORTRAIT_DATA` JavaScript object near the top of the `\u003Cbody>`. Replace the ENTIRE `PORTRAIT_DATA` object with the computed data. The structure is:\n\n```javascript\nconst PORTRAIT_DATA = {\n  meta: {\n    username: \"...\",         \u002F\u002F infer from system username or ask\n    generatedAt: \"...\",      \u002F\u002F today's date\n    messageCount: 0,         \u002F\u002F actual count of sampled messages\n    dateRange: \"...\",        \u002F\u002F earliest to latest message date\n    confidence: \"...\",       \u002F\u002F \"Early Sketch\" \u002F \"Clear Portrait\" \u002F \"Deep Portrait\"\n    lang: \"...\"              \u002F\u002F \"zh\" if user's messages are predominantly Chinese, otherwise \"en\"\n  },\n  personality: {\n    mbtiType: \"XXXX\",\n    axes: {\n      EI: { score: 0, letter: \"X\", label: \"...\" },\n      SN: { score: 0, letter: \"X\", label: \"...\" },\n      TF: { score: 0, letter: \"X\", label: \"...\" },\n      JP: { score: 0, letter: \"X\", label: \"...\" }\n    },\n    summary: \"...\"\n  },\n  radar: {\n    technicalDepth: 0,\n    technicalBreadth: 0,\n    communication: 0,\n    decisionSpeed: 0,\n    collaboration: 0,\n    creativity: 0\n  },\n  rating: {\n    tier: \"...\",\n    label: \"...\",             \u002F\u002F Chinese label: 夯爆了\u002F夯\u002F人上人\u002FNPC\u002F拉\u002F拉完了\n    labelEn: \"...\",           \u002F\u002F English label: Legendary\u002FElite\u002FAbove Average\u002FNPC\u002FBelow Average\u002FInactive\n    emoji: \"...\",\n    color: \"...\",\n    reason: \"...\",\n    strengthHighlights: [],\n    growthAreas: []\n  },\n  famousMatch: {\n    technical: {\n      name: \"...\", emoji: \"...\",\n      dimension: \"Technical Spirit\", dimensionZh: \"技术灵魂\",\n      reason: \"...\", sharedTraits: []\n    },\n    strategic: {\n      name: \"...\", emoji: \"...\",\n      dimension: \"Strategic Mind\", dimensionZh: \"思维内核\",\n      reason: \"...\", sharedTraits: []\n    },\n    communication: {\n      name: \"...\", emoji: \"...\",\n      dimension: \"Communication Soul\", dimensionZh: \"表达人格\",\n      reason: \"...\", sharedTraits: []\n    }\n  },\n  communication: {\n    languages: {},\n    directnessScore: 0,\n    avgMessageLength: 0,\n    topKeywords: [],\n    questionRatio: 0\n  },\n  technical: {\n    domains: {},\n    topTools: []\n  },\n  workRhythm: {\n    hourlyActivity: [\u002F* 24 integers, one per hour *\u002F],\n    sessionPattern: \"...\",\n    avgSessionLength: \"...\",\n    projectCount: 0\n  },\n  quotes: [\n    { text: \"...\", context: \"...\" }\n    \u002F\u002F 3-5 representative quotes\n  ]\n};\n```\n\n**IMPORTANT:**\n- Replace ONLY the `PORTRAIT_DATA` object. Do not modify any other part of the template.\n- All values must be valid JavaScript (strings quoted, arrays bracketed, no trailing commas).\n- The rest of the template's JavaScript will automatically render everything from this object.\n- Keep quote texts under 100 characters. Anonymize any sensitive data (API keys, tokens, passwords, personal URLs).\n\n## Step 5: Generate persona skill\n\nRead the persona skill template:\n→ `references\u002Fpersona-skill-template.md`\n\nGenerate a **multi-file** persona skill. Read the template for the full structure:\n→ `references\u002Fpersona-skill-template.md`\n\nCreate the following files under `~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002F`:\n\n```\nme\u002F\n├── SKILL.md                        # Entry point (~80 lines)\n├── portrait-meta.json              # Timestamps for incremental updates\n└── references\u002F\n    ├── thinking-patterns.md        # From: technical depth + decision patterns\n    ├── decision-framework.md       # From: decision patterns + collaboration style\n    ├── communication-style.md      # From: communication style analysis\n    ├── engineering-philosophy.md   # From: technical depth + breadth + anti-patterns\n    └── mindset-markers.md          # Abstracted attitudes (NO raw quotes)\n```\n\n**`portrait-meta.json` is critical** — it stores `lastMessageIndex` (line count of history.jsonl at analysis time) and `updatedAt` timestamp. These enable the **update** subcommand to do incremental analysis later.\n\nCreate directories if they don't exist. `me\u002F` is reserved for the user's own persona.\n\n## Step 6: Write output files\n\nWrite the following files to the current working directory:\n\n**1. HTML Portrait:**\n```\nvibe-portrait-YYYY-MM-DD.html\n```\n\n**2. Portrait Image (auto-export):**\n\nAfter writing the HTML, attempt to generate a PNG screenshot of the portrait. Try these methods in order:\n\n- **If Playwright MCP is available** (check for `playwright_navigate` and `playwright_screenshot` tools):\n  1. Navigate to the local HTML file: `file:\u002F\u002F\u002Fabsolute\u002Fpath\u002Fto\u002Fvibe-portrait-YYYY-MM-DD.html`\n  2. Wait for charts to render (2-3 seconds)\n  3. Take a full-page screenshot, save as `vibe-portrait-YYYY-MM-DD.png`\n\n- **If Playwright is not available**, try bash:\n  ```bash\n  # Try with npx playwright if installed\n  npx --yes playwright screenshot --full-page \"file:\u002F\u002F$(pwd)\u002Fvibe-portrait-YYYY-MM-DD.html\" \"vibe-portrait-YYYY-MM-DD.png\" 2>\u002Fdev\u002Fnull\n  ```\n\n- **If neither works**, skip silently. Tell the user they can use the \"Export as Image\" button on the HTML page instead.\n\n**3. Analysis JSON** (for sync\u002Fmerge):\n```\nvibe-portrait-analysis-YYYY-MM-DD.json\n```\n\nContents:\n```json\n{\n  \"version\": \"1.0\",\n  \"machine\": \"\u003Chostname from `hostname` command>\",\n  \"exportedAt\": \"YYYY-MM-DD\",\n  \"sourceFiles\": [\"~\u002F.claude\u002Fhistory.jsonl\"],\n  \"messageCount\": 200,\n  \"portraitData\": { \u002F* the full PORTRAIT_DATA object *\u002F }\n}\n```\n\n**3. Persona skill** (also written in Step 5):\n```\n~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002FSKILL.md\n```\n\n## Step 7: Sync to portrait repo\n\nAfter writing all output files, ask the user:\n\n> **Sync to your portrait repo?**\n>\n> 1. **Yes — create a new repo** (first time)\n> 2. **Yes — push to existing repo**\n> 3. **No — keep everything local**\n\n### Option 1: Create new portrait repo\n\n1. Check `gh auth status`. If not authenticated, tell user to run `! gh auth login`.\n2. Copy `repo-template\u002F` contents into a temp directory.\n3. Copy the generated analysis JSON, HTML portrait, and persona skill into the temp directory under `analysis\u002F`, `portraits\u002F`, and `me\u002F` respectively.\n4. **Fill in the README.md** — the template contains `{{PLACEHOLDER}}` markers. Read `repo-template\u002FREADME.md` (see the file for the full list of placeholders) and replace them all with the user's analysis data. Key formatting rules:\n   - Badge lists → inline code: `` `item1` `item2` ``\n   - Domain bars → Unicode blocks: `**ML\u002FAI** ██████████████░░░░░░ 30%` (20 chars wide, `█` filled, `░` empty)\n   - Quotes → blockquotes: `> \"text\" — *context*`\n   - Language breakdown → `\"Chinese 65% · English 30% · Mixed 5%\"`\n5. `git init && git add -A && git commit -m \"Initial portrait\"`\n6. `gh repo create my-vibe-portrait --private --source=. --push`\n7. Save repo URL: `gh repo view --json url -q '.url' > ~\u002F.vibe-portrait-repo`\n\n### Option 2: Push to existing repo\n\n1. Read repo URL from `~\u002F.vibe-portrait-repo`.\n2. Clone (or pull if already cached at `~\u002F.cache\u002Fvibe-portrait-sync`).\n3. Copy the latest analysis JSON, HTML, and persona skill into the repo directory.\n4. Regenerate `README.md` using the same placeholder-filling process as Option 1.\n5. `git add -A && git commit -m \"Update portrait from $(hostname)\" && git push`\n\n### Option 3: Keep local\n\nDo nothing. Files are already written locally.\n\n## Step 8: Present results\n\nTell the user:\n1. Where the HTML portrait was saved (open in browser)\n2. Where the persona skill was installed (`~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002F`)\n3. Whether the portrait repo was synced (and the repo URL if applicable)\n4. A brief summary of key findings (MBTI type, rating, famous match)\n5. How to use the persona: **\"think like me\"** or **\"像我一样思考\"**\n6. How to sync from another machine: install vibe-portrait there, run it, choose \"push to existing repo\"\n\n## Confidence levels\n\nBased on total non-empty messages found:\n- **\u003C 20**: Do not generate. Tell user to accumulate more history.\n- **20-50**: Generate with `confidence: \"Early Sketch\"`. Warn that results are approximate.\n- **50-200**: Generate with `confidence: \"Clear Portrait\"`.\n- **200+**: Generate with `confidence: \"Deep Portrait\"`.\n\n## Output contract\n\nThis skill produces THREE outputs:\n\n### 1. HTML Portrait (`vibe-portrait-YYYY-MM-DD.html`)\n- Single self-contained file (only external dependencies: Tailwind CDN + Chart.js CDN)\n- Opens correctly in any modern browser\n- Contains all analysis data in the `PORTRAIT_DATA` JavaScript object\n- Includes actual quotes (anonymized for sensitive content)\n\n### 2. Persona Skill (`~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002FSKILL.md`)\n- Valid Claude Code \u002F Codex skill file with proper YAML frontmatter\n- Under 200 lines, dense and actionable\n- Activates on \"think like me\" \u002F \"像我一样思考\" and similar phrases\n- Captures thinking patterns, decision framework, communication style, engineering philosophy\n\n### 3. Analysis JSON (`vibe-portrait-analysis-YYYY-MM-DD.json`)\n- Portable data file for cross-machine syncing\n- Contains full PORTRAIT_DATA plus machine hostname and export metadata\n- Synced to portrait repo (if configured) or kept local\n\n### 4. Portrait Repo (optional, private GitHub repo)\n- Created via `gh repo create my-vibe-portrait --private`\n- Stores persona skill, analysis JSONs from all machines, and latest portrait\n- Repo URL cached in `~\u002F.vibe-portrait-repo` for future syncs\n- Template provided in `repo-template\u002F` within this skill\n\n## Privacy rules (CRITICAL — read before generating ANY output)\n\nAll outputs of VibePortrait (HTML, persona skill, analysis JSON, repo README) are potentially shareable. Apply these rules to **every file you write**:\n\n### What MUST be redacted\n- **API keys, tokens, passwords, secrets** → replace with `[REDACTED]`\n- **URLs containing credentials** (tokens in query strings, auth headers) → redact the credential portion\n- **File paths exposing personal info** (`\u002Fhome\u002Fusername\u002F`, `\u002FUsers\u002Frealname\u002F`) → replace with `~\u002F` or `\u003Chome>\u002F`\n- **Email addresses, phone numbers, IPs** → redact unless the user explicitly consents\n- **Private repo URLs, internal hostnames** → redact\n\n### What MUST NOT appear in persona skills\n- **No verbatim chat messages.** Persona reference files must contain only abstracted personality descriptions, never raw conversation text.\n- **No project names, company names, or colleague names** from the user's conversations\n- **No financial data** (account balances, trading strategies, PnL numbers)\n- **No credentials or deployment details** (server IPs, Docker configs, database names)\n\n### What is OK\n- Abstracted personality traits, thinking patterns, communication style descriptions\n- Generic domain labels (\"ML\u002FAI\", \"Web\", \"Systems\u002FInfra\")\n- Tool\u002Fframework names (public knowledge: \"PyTorch\", \"Docker\", \"React\")\n- MBTI scores, radar scores, rating tier (all computed, not raw data)\n\n### Before pushing to git repo\nScan all files being committed for the redaction patterns above. If any slip through, fix before pushing.\n\n## Anti-patterns\n\n- **Do not fabricate data.** If you can't determine a signal, use a neutral score (50).\n- **Do not read files beyond conversation history.** No source code, private documents, SSH keys.\n- **Do not skip the template.** Always use `templates\u002Fportrait.html` as the base.\n- **Do not read all history lines in Quick mode.** Use the sampling strategy.\n\n## Reference files\n\n`references\u002Fanalysis-framework.md` · `references\u002Fmbti-mapping.md` · `references\u002Ffamous-matching.md` · `references\u002Frating-rubric.md` · `references\u002Fpersona-skill-template.md` · `templates\u002Fportrait.html` · `repo-template\u002F`\n\n## Update (incremental analysis)\n\nWhen the user says \"update my portrait\" \u002F \"更新我的画像\":\n\n1. Read `~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002Fme\u002Fportrait-meta.json`.\n   - If it doesn't exist, tell the user to run a full generate first.\n2. Get `lastMessageIndex` from the meta file.\n3. Count current lines in `~\u002F.claude\u002Fhistory.jsonl`.\n4. If no new messages since last analysis, tell the user \"Portrait is up to date.\"\n5. Otherwise, read only the **new messages** (from line `lastMessageIndex+1` to end).\n6. Analyze the new messages across the same 6 dimensions.\n7. **Merge** with existing analysis: weighted average of scores by message count (old weight = `totalMessagesAnalyzed`, new weight = new message count).\n8. Regenerate: HTML portrait, all persona reference files, portrait-meta.json (update `updatedAt`, `lastMessageIndex`, `totalMessagesAnalyzed`).\n9. If portrait repo exists, push the update.\n\nThis is much cheaper than a full re-analysis — typically only processes dozens of new messages.\n\n---\n\n## Persona Management\n\n### Install from GitHub\n\nTrigger: \"install persona from \\\u003Curl\\>\"\n\n```bash\nTEMP_DIR=$(mktemp -d)\ngit clone --depth 1 \"\u003Curl>\" \"$TEMP_DIR\"\nPERSON_ID=$(grep -m1 'personaId' \"$TEMP_DIR\u002Fme\u002Fportrait-meta.json\" | sed 's\u002F.*: *\"\\(.*\\)\".*\u002F\\1\u002F' )\n# Fallback: infer from GitHub username\nTARGET=~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002F$PERSON_ID\ncp -R \"$TEMP_DIR\u002Fme\" \"$TARGET\"\nrm -rf \"$TEMP_DIR\"\n```\n\nConfirm: where installed + how to invoke (`think like {person-id}`).\n\n### List\n\nTrigger: \"list personas\" \u002F \"我安装了哪些人格\"\n\n```bash\nls ~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002F\n```\n\nFor each directory, read `portrait-meta.json` and show: name, MBTI type, rating, last updated.\n\n### Remove\n\nTrigger: \"remove persona \\\u003Cid\\>\"\n\n```bash\nrm -rf ~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002F\u003Cid>\n```\n\nConfirm before deleting. Refuse to delete `me\u002F` — tell user to run generate to overwrite instead.\n\n### Directory layout\n\n```\n~\u002F.claude\u002Fskills\u002Fvibe-portrait-personas\u002F\n├── me\u002F                        ← yours (multi-file, auto-generated)\n│   ├── SKILL.md\n│   ├── portrait-meta.json\n│   └── references\u002F\n│       ├── thinking-patterns.md\n│       ├── decision-framework.md\n│       ├── communication-style.md\n│       ├── engineering-philosophy.md\n│       └── mindset-markers.md\n├── jane-doe\u002F                  ← installed from GitHub\n│   ├── SKILL.md\n│   ├── portrait-meta.json\n│   └── references\u002F...\n└── zhuge-liang\u002F               ← installed from community\n    └── ...\n",{"name":5,"description":7},"main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait",491,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-14 14:06:13",{"source":8,"skillName":5,"commands":61,"prompt":71,"zipAvailable":72},[62,65,68],{"tool":63,"command":64},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill vibe-portrait",{"tool":66,"command":67},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill vibe-portrait",{"tool":69,"command":70},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill vibe-portrait","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002FSKILL.md",true,[74,80,84,88,92,96,100,106,110,113,116,120],{"path":75,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":78,"sourceHtmlUrl":79},"agents\u002Fopenai.yaml","REFERENCE","参考文档",306,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Fagents\u002Fopenai.yaml",{"path":81,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":82,"sourceHtmlUrl":83},"references\u002Fanalysis-framework.md",5906,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Freferences\u002Fanalysis-framework.md",{"path":85,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":86,"sourceHtmlUrl":87},"references\u002Ffamous-matching.md",3880,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Freferences\u002Ffamous-matching.md",{"path":89,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":90,"sourceHtmlUrl":91},"references\u002Fmbti-mapping.md",4935,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Freferences\u002Fmbti-mapping.md",{"path":93,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":94,"sourceHtmlUrl":95},"references\u002Fpersona-skill-template.md",7038,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Freferences\u002Fpersona-skill-template.md",{"path":97,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":98,"sourceHtmlUrl":99},"references\u002Frating-rubric.md",7129,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Freferences\u002Frating-rubric.md",{"path":101,"resourceType":102,"resourceTypeDesc":103,"sizeBytes":104,"sourceHtmlUrl":105},"repo-template\u002F.gitignore","OTHER","其他",25,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Frepo-template\u002F.gitignore",{"path":107,"resourceType":102,"resourceTypeDesc":103,"sizeBytes":108,"sourceHtmlUrl":109},"repo-template\u002Fanalysis\u002F.gitkeep",0,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Frepo-template\u002Fanalysis\u002F.gitkeep",{"path":111,"resourceType":102,"resourceTypeDesc":103,"sizeBytes":108,"sourceHtmlUrl":112},"repo-template\u002Fme\u002F.gitkeep","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Frepo-template\u002Fme\u002F.gitkeep",{"path":114,"resourceType":102,"resourceTypeDesc":103,"sizeBytes":108,"sourceHtmlUrl":115},"repo-template\u002Fportraits\u002F.gitkeep","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Frepo-template\u002Fportraits\u002F.gitkeep",{"path":117,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":118,"sourceHtmlUrl":119},"repo-template\u002FREADME.md",2738,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Frepo-template\u002FREADME.md",{"path":121,"resourceType":76,"resourceTypeDesc":77,"sizeBytes":122,"sourceHtmlUrl":123},"templates\u002Fportrait.html",45917,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fdadwadw233\u002FVibePortrait\u002Fskills\u002Fvibe-portrait\u002Ftemplates\u002Fportrait.html",[125,135,143,151,159,167],{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":129,"licenseFileUrl":130,"sourceHtmlUrl":131,"categories":132,"keywords":133,"publishedAt":134},"01M7DV332YQMGD8WTRJ9DZQXTB","worktable-review-with-annotations","Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.","MIT","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-review-with-annotations\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":136,"name":137,"nameZh":6,"description":138,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":129,"licenseFileUrl":139,"sourceHtmlUrl":140,"categories":141,"keywords":142,"publishedAt":134},"01M7DV2YGHCZVBEB1AHYB6CGHE","worktable-find-and-synthesize-context","Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-find-and-synthesize-context\u002FSKILL.md",[],[],{"skillNo":144,"name":145,"nameZh":6,"description":146,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":129,"licenseFileUrl":147,"sourceHtmlUrl":148,"categories":149,"keywords":150,"publishedAt":134},"01M7DV2TQN1H11ME2SAF9MXN7V","worktable-create-or-update-docs","Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-update-docs\u002FSKILL.md",[],[],{"skillNo":152,"name":153,"nameZh":6,"description":154,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":129,"licenseFileUrl":155,"sourceHtmlUrl":156,"categories":157,"keywords":158,"publishedAt":134},"01M7DV2NE452NEA8TXATKSNAJH","worktable-create-or-manage-records","Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-records\u002FSKILL.md",[],[],{"skillNo":160,"name":161,"nameZh":6,"description":162,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":129,"licenseFileUrl":163,"sourceHtmlUrl":164,"categories":165,"keywords":166,"publishedAt":134},"01M7DV2HENRBR2KSCVM064WH0Y","worktable-create-or-manage-interactive-html","Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-create-or-manage-interactive-html\u002FSKILL.md",[],[],{"skillNo":168,"name":169,"nameZh":6,"description":170,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":129,"licenseFileUrl":171,"sourceHtmlUrl":172,"categories":173,"keywords":174,"publishedAt":134},"01M7DV2CADR5R7RQFFYPNAGR9Y","worktable-collaborate-threads","Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fworktable\u002Fworktable-plugin\u002Fskills\u002Fworktable-collaborate-threads\u002FSKILL.md",[],[]]