[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6BVW4039XX633V95QMRVHT5":3,"skill-same:01M6BVW4039XX633V95QMRVHT5":77},{"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":41,"linkOnly":42,"markdownContent":43,"markdownBody":44,"frontMatter":45,"refName":48,"commitSha":49,"skillDirectory":50,"forksCount":51,"repositoryHtmlUrl":52,"repositoryHomepageUrl":6,"ownerHtmlUrl":53,"ownerBlogUrl":54,"ownerDescription":55,"updatedAt":56,"install":57,"resources":70},"01M6BVW4039XX633V95QMRVHT5","opencode-memory",null,"Browse and recall OpenCode local memory stored on the user's machine: local sessions, plans, conversations, prompt history, and project context. Use immediately when the user asks to check history, previous sessions, past chats, what did we do before, last time, check plans, session history, recall, memory, remember, prior work, previous context, or have we done this before. Auto-trigger proactively when resuming work, continuing a project, referencing prior decisions, debugging repeated issues, revisiting earlier plans, or any follow-up where earlier OpenCode context may help. This means OpenCode local history\u002Ffiles specifically, not ChatGPT\u002FClaude cloud history, generic web search, or unrelated product memory systems. Do NOT use for fresh tasks with no relevant history, or when current files\u002Fgit already answer the question.","ajoslin\u002Fdot","ajoslin","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F1265995?v=4","01M6BVTK48NGBP9V32M72JAX0V","Andrew Joslin","01M6BVTK8SSEMZK27ZF7PX0PV9",24,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fajoslin\u002Fdot\u002Fblob\u002Fmaster\u002Fconfig\u002Fopencode\u002Fskills\u002Fopencode-memory\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40],"opencode memory","memory","issues","web search","recall","browse","cloud","search","chatgpt","product","git","help","tasks","debugging","remember","sessions","web","2026-11-10 10:12:30",false,"---\nname: opencode-memory\ndescription: >\n  Browse and recall OpenCode local memory stored on the user's machine: local\n  sessions, plans, conversations, prompt history, and project context. Use\n  immediately when the user asks to check history, previous sessions, past\n  chats, what did we do before, last time, check plans, session history,\n  recall, memory, remember, prior work, previous context, or have we done this\n  before. Auto-trigger proactively when resuming work, continuing a project,\n  referencing prior decisions, debugging repeated issues, revisiting earlier\n  plans, or any follow-up where earlier OpenCode context may help. This means\n  OpenCode local history\u002Ffiles specifically, not ChatGPT\u002FClaude cloud history,\n  generic web search, or unrelated product memory systems. Do NOT use for fresh\n  tasks with no relevant history, or when current files\u002Fgit already answer the\n  question.\nlicense: Apache-2.0\ncompatibility: opencode\n---\n\n# OpenCode Memory Browser\n\nLightweight, read-only access to your local OpenCode history. No injection, no bloat — just the ability to look things up when it would help.\n\nThis skill is specifically about OpenCode data stored on the local machine. It is not for ChatGPT history, Claude cloud history, generic browser history, or external memory products.\n\nAll data lives in a local SQLite database and plain files. You query them directly using `sqlite3` via bash. No bundled scripts or external dependencies needed.\n\n## When to Use\n\n### Auto-trigger (agent decides)\n\n- You are resuming work on a project and suspect prior sessions exist.\n- The user references something done previously (\"we did this before\", \"last time\", \"that plan we made\").\n- A recurring issue suggests checking if it was encountered before.\n- The user asks about the state of plans, past decisions, or previous approaches.\n- You need context that might exist in history but is not in the current session.\n\n### User-triggered (explicit request)\n\n- \"Check my history\"\n- \"What did we do in the last session?\"\n- \"Show me my plans\"\n- \"Search for when we discussed X\"\n- \"What projects have I worked on?\"\n- \"Look at previous conversations about Y\"\n\n### Do NOT use when\n\n- The task is clearly brand new with no relevant history.\n- Fresh repo context (files, git log) is sufficient.\n- The user explicitly says they don't care about prior work.\n\n## Storage Locations\n\n```\nDatabase:       ${XDG_DATA_HOME:-$HOME\u002F.local\u002Fshare}\u002Fopencode\u002Fopencode.db\nPlans:          ${XDG_DATA_HOME:-$HOME\u002F.local\u002Fshare}\u002Fopencode\u002Fplans\u002F*.md\nSession diffs:  ${XDG_DATA_HOME:-$HOME\u002F.local\u002Fshare}\u002Fopencode\u002Fstorage\u002Fsession_diff\u002F\u003Csession-id>.json\nPrompt history: ${XDG_STATE_HOME:-$HOME\u002F.local\u002Fstate}\u002Fopencode\u002Fprompt-history.jsonl\n```\n\nThe database path respects `$XDG_DATA_HOME` if set (default: `~\u002F.local\u002Fshare`).\n\n## Database Schema (what matters)\n\n- **project** — `id` (text PK), `worktree` (path), `name` (often NULL, derive from worktree basename)\n- **session** — `id` (text, e.g. `ses_xxx`), `project_id` (FK), `parent_id` (NULL = main session, set = subagent), `title`, `summary`, `time_created`, `time_updated`\n- **message** — `id`, `session_id` (FK), `data` (JSON with `$.role` = `\"user\"` or `\"assistant\"`), `time_created`\n- **part** — `id`, `message_id` (FK), `session_id` (FK), `data` (JSON with `$.type` = `\"text\"` and `$.text` = content)\n\nTimestamps are Unix milliseconds. Use `datetime(col\u002F1000, 'unixepoch', 'localtime')` to display them.\n\n## Ready-to-Use Queries\n\nAll queries use `sqlite3` in read-only mode. Always run via bash.\n\n**Shorthand used below:**\n```\nDATA_ROOT=\"${XDG_DATA_HOME:-$HOME\u002F.local\u002Fshare}\u002Fopencode\"\nSTATE_ROOT=\"${XDG_STATE_HOME:-$HOME\u002F.local\u002Fstate}\u002Fopencode\"\nDB=\"$DATA_ROOT\u002Fopencode.db\"\nDB_URI=\"file:${DB}?mode=ro\"\n```\n\n### Quick summary\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT 'projects', COUNT(*) FROM project\n  UNION ALL SELECT 'sessions (main)', COUNT(*) FROM session WHERE parent_id IS NULL\n  UNION ALL SELECT 'sessions (total)', COUNT(*) FROM session\n  UNION ALL SELECT 'messages', COUNT(*) FROM message\n  UNION ALL SELECT 'todos', COUNT(*) FROM todo;\n\"\n```\n\n### List projects\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT\n    COALESCE(p.name, CASE WHEN p.worktree = '\u002F' THEN '(global)' ELSE REPLACE(p.worktree, RTRIM(p.worktree, REPLACE(p.worktree, '\u002F', '')), '') END) AS name,\n    p.worktree,\n    (SELECT COUNT(*) FROM session s WHERE s.project_id = p.id AND s.parent_id IS NULL) AS sessions\n  FROM project p\n  ORDER BY p.time_updated DESC\n  LIMIT 10;\n\"\n```\n\n### List recent sessions\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT\n    s.id,\n    COALESCE(s.title, 'untitled') AS title,\n    COALESCE(p.name, CASE WHEN p.worktree = '\u002F' THEN '(global)' ELSE REPLACE(p.worktree, RTRIM(p.worktree, REPLACE(p.worktree, '\u002F', '')), '') END) AS project,\n    datetime(s.time_updated\u002F1000, 'unixepoch', 'localtime') AS updated,\n    (SELECT COUNT(*) FROM message m WHERE m.session_id = s.id) AS msgs\n  FROM session s\n  LEFT JOIN project p ON p.id = s.project_id\n  WHERE s.parent_id IS NULL\n  ORDER BY s.time_updated DESC\n  LIMIT 10;\n\"\n```\n\n### Sessions for a specific project\n\nReplace the worktree path with the actual project path:\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT s.id, COALESCE(s.title, 'untitled'),\n    datetime(s.time_updated\u002F1000, 'unixepoch', 'localtime')\n  FROM session s\n  JOIN project p ON p.id = s.project_id\n  WHERE p.worktree = '\u002Fpath\u002Fto\u002Fproject'\n    AND s.parent_id IS NULL\n  ORDER BY s.time_updated DESC\n  LIMIT 10;\n\"\n```\n\nTo find the worktree for the current directory: `git rev-parse --show-toplevel`\n\n### Read messages from a session\n\nReplace the session ID:\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT\n    json_extract(m.data, '$.role') AS role,\n    datetime(m.time_created\u002F1000, 'unixepoch', 'localtime') AS time,\n    GROUP_CONCAT(json_extract(p.data, '$.text'), char(10)) AS text\n  FROM message m\n  LEFT JOIN part p ON p.message_id = m.id\n    AND json_extract(p.data, '$.type') = 'text'\n  WHERE m.session_id = 'SESSION_ID_HERE'\n  GROUP BY m.id\n  ORDER BY m.time_created ASC\n  LIMIT 50;\n\"\n```\n\n### Search across all conversations\n\nReplace the search term:\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT\n    s.id AS session_id,\n    COALESCE(s.title, 'untitled') AS title,\n    json_extract(m.data, '$.role') AS role,\n    datetime(m.time_created\u002F1000, 'unixepoch', 'localtime') AS time,\n    substr(json_extract(p.data, '$.text'), 1, 200) AS snippet\n  FROM part p\n  JOIN message m ON m.id = p.message_id\n  JOIN session s ON s.id = m.session_id\n  WHERE s.parent_id IS NULL\n    AND json_extract(p.data, '$.type') = 'text'\n    AND json_extract(p.data, '$.text') LIKE '%SEARCH_TERM%'\n  ORDER BY m.time_created DESC\n  LIMIT 10;\n\"\n```\n\n### List saved plans\n\n```bash\nls -lt \"$DATA_ROOT\"\u002Fplans\u002F*.md 2>\u002Fdev\u002Fnull | head -20\n```\n\nTo read a specific plan:\n\n```bash\ncat \"$DATA_ROOT\"\u002Fplans\u002FFILENAME.md\n```\n\n### Show recent prompt history\n\n```bash\ntail -20 \"$STATE_ROOT\"\u002Fprompt-history.jsonl\n```\n\nEach line is a JSON object. The user's input is typically in the `input` or `text` field.\n\n## Workflow\n\n### Quick recall (most common)\n\n1. Run the **summary** query to see what's available.\n2. If you need sessions for the current project, get the worktree with `git rev-parse --show-toplevel`, then run the **project sessions** query.\n3. If you need a specific topic, run the **search** query.\n4. If you need full conversation detail, run the **messages** query with the session ID.\n\n### Plan review\n\n1. List plans with `ls -lt \"$DATA_ROOT\"\u002Fplans\u002F*.md`.\n2. Read a plan with `cat \"$DATA_ROOT\"\u002Fplans\u002F\u003Cfilename>.md`.\n\n### Deep investigation\n\n1. Run **projects** to see all tracked repos.\n2. Run **sessions** for a specific project.\n3. Run **messages** for full conversation content.\n4. Cross-reference with **search** across all projects.\n\n## Critical Rules\n\n1. **Read-only.** Never write to or modify the database or any OpenCode files.\n2. **Use bash + sqlite3.** Do not try to read `opencode.db` with the Read tool — it is a binary file. Always query via `sqlite3` in bash.\n3. **Don't dump everything.** Use `LIMIT` and `LIKE` to keep output focused. The database can contain tens of thousands of messages.\n4. **Summarize for the user.** After retrieving data, distill the relevant parts. Don't paste raw query output.\n5. **Respect privacy.** Session history may contain sensitive data. Only surface what is relevant to the current task.\n6. **Set path variables first.** At the start of any memory lookup, set `DATA_ROOT`, `STATE_ROOT`, `DB`, and `DB_URI` exactly as shown above so the commands work on XDG and non-XDG setups and keep SQLite access read-only.\n\n## Fallback: Web UI\n\nIf the user needs visual dashboards or a browsable interface:\n\n1. Check if OpenCode web is running: `curl -s http:\u002F\u002F127.0.0.1:4096\u002Fapi\u002Fhealth 2>\u002Fdev\u002Fnull || echo \"not running\"`\n2. If running, direct the user to `http:\u002F\u002F127.0.0.1:4096`.\n3. If not running, suggest `opencode web`.\n4. Note: `opencode.local` only works with mDNS enabled (`opencode web --mdns`). Don't assume it exists.\n\n## Deep Reference\n\nSee [references\u002Fstorage-format.md](references\u002Fstorage-format.md) for the full storage layout, all table schemas, and additional query examples.\n","# OpenCode Memory Browser\n\nLightweight, read-only access to your local OpenCode history. No injection, no bloat — just the ability to look things up when it would help.\n\nThis skill is specifically about OpenCode data stored on the local machine. It is not for ChatGPT history, Claude cloud history, generic browser history, or external memory products.\n\nAll data lives in a local SQLite database and plain files. You query them directly using `sqlite3` via bash. No bundled scripts or external dependencies needed.\n\n## When to Use\n\n### Auto-trigger (agent decides)\n\n- You are resuming work on a project and suspect prior sessions exist.\n- The user references something done previously (\"we did this before\", \"last time\", \"that plan we made\").\n- A recurring issue suggests checking if it was encountered before.\n- The user asks about the state of plans, past decisions, or previous approaches.\n- You need context that might exist in history but is not in the current session.\n\n### User-triggered (explicit request)\n\n- \"Check my history\"\n- \"What did we do in the last session?\"\n- \"Show me my plans\"\n- \"Search for when we discussed X\"\n- \"What projects have I worked on?\"\n- \"Look at previous conversations about Y\"\n\n### Do NOT use when\n\n- The task is clearly brand new with no relevant history.\n- Fresh repo context (files, git log) is sufficient.\n- The user explicitly says they don't care about prior work.\n\n## Storage Locations\n\n```\nDatabase:       ${XDG_DATA_HOME:-$HOME\u002F.local\u002Fshare}\u002Fopencode\u002Fopencode.db\nPlans:          ${XDG_DATA_HOME:-$HOME\u002F.local\u002Fshare}\u002Fopencode\u002Fplans\u002F*.md\nSession diffs:  ${XDG_DATA_HOME:-$HOME\u002F.local\u002Fshare}\u002Fopencode\u002Fstorage\u002Fsession_diff\u002F\u003Csession-id>.json\nPrompt history: ${XDG_STATE_HOME:-$HOME\u002F.local\u002Fstate}\u002Fopencode\u002Fprompt-history.jsonl\n```\n\nThe database path respects `$XDG_DATA_HOME` if set (default: `~\u002F.local\u002Fshare`).\n\n## Database Schema (what matters)\n\n- **project** — `id` (text PK), `worktree` (path), `name` (often NULL, derive from worktree basename)\n- **session** — `id` (text, e.g. `ses_xxx`), `project_id` (FK), `parent_id` (NULL = main session, set = subagent), `title`, `summary`, `time_created`, `time_updated`\n- **message** — `id`, `session_id` (FK), `data` (JSON with `$.role` = `\"user\"` or `\"assistant\"`), `time_created`\n- **part** — `id`, `message_id` (FK), `session_id` (FK), `data` (JSON with `$.type` = `\"text\"` and `$.text` = content)\n\nTimestamps are Unix milliseconds. Use `datetime(col\u002F1000, 'unixepoch', 'localtime')` to display them.\n\n## Ready-to-Use Queries\n\nAll queries use `sqlite3` in read-only mode. Always run via bash.\n\n**Shorthand used below:**\n```\nDATA_ROOT=\"${XDG_DATA_HOME:-$HOME\u002F.local\u002Fshare}\u002Fopencode\"\nSTATE_ROOT=\"${XDG_STATE_HOME:-$HOME\u002F.local\u002Fstate}\u002Fopencode\"\nDB=\"$DATA_ROOT\u002Fopencode.db\"\nDB_URI=\"file:${DB}?mode=ro\"\n```\n\n### Quick summary\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT 'projects', COUNT(*) FROM project\n  UNION ALL SELECT 'sessions (main)', COUNT(*) FROM session WHERE parent_id IS NULL\n  UNION ALL SELECT 'sessions (total)', COUNT(*) FROM session\n  UNION ALL SELECT 'messages', COUNT(*) FROM message\n  UNION ALL SELECT 'todos', COUNT(*) FROM todo;\n\"\n```\n\n### List projects\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT\n    COALESCE(p.name, CASE WHEN p.worktree = '\u002F' THEN '(global)' ELSE REPLACE(p.worktree, RTRIM(p.worktree, REPLACE(p.worktree, '\u002F', '')), '') END) AS name,\n    p.worktree,\n    (SELECT COUNT(*) FROM session s WHERE s.project_id = p.id AND s.parent_id IS NULL) AS sessions\n  FROM project p\n  ORDER BY p.time_updated DESC\n  LIMIT 10;\n\"\n```\n\n### List recent sessions\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT\n    s.id,\n    COALESCE(s.title, 'untitled') AS title,\n    COALESCE(p.name, CASE WHEN p.worktree = '\u002F' THEN '(global)' ELSE REPLACE(p.worktree, RTRIM(p.worktree, REPLACE(p.worktree, '\u002F', '')), '') END) AS project,\n    datetime(s.time_updated\u002F1000, 'unixepoch', 'localtime') AS updated,\n    (SELECT COUNT(*) FROM message m WHERE m.session_id = s.id) AS msgs\n  FROM session s\n  LEFT JOIN project p ON p.id = s.project_id\n  WHERE s.parent_id IS NULL\n  ORDER BY s.time_updated DESC\n  LIMIT 10;\n\"\n```\n\n### Sessions for a specific project\n\nReplace the worktree path with the actual project path:\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT s.id, COALESCE(s.title, 'untitled'),\n    datetime(s.time_updated\u002F1000, 'unixepoch', 'localtime')\n  FROM session s\n  JOIN project p ON p.id = s.project_id\n  WHERE p.worktree = '\u002Fpath\u002Fto\u002Fproject'\n    AND s.parent_id IS NULL\n  ORDER BY s.time_updated DESC\n  LIMIT 10;\n\"\n```\n\nTo find the worktree for the current directory: `git rev-parse --show-toplevel`\n\n### Read messages from a session\n\nReplace the session ID:\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT\n    json_extract(m.data, '$.role') AS role,\n    datetime(m.time_created\u002F1000, 'unixepoch', 'localtime') AS time,\n    GROUP_CONCAT(json_extract(p.data, '$.text'), char(10)) AS text\n  FROM message m\n  LEFT JOIN part p ON p.message_id = m.id\n    AND json_extract(p.data, '$.type') = 'text'\n  WHERE m.session_id = 'SESSION_ID_HERE'\n  GROUP BY m.id\n  ORDER BY m.time_created ASC\n  LIMIT 50;\n\"\n```\n\n### Search across all conversations\n\nReplace the search term:\n\n```bash\nsqlite3 \"$DB_URI\" \"\n  SELECT\n    s.id AS session_id,\n    COALESCE(s.title, 'untitled') AS title,\n    json_extract(m.data, '$.role') AS role,\n    datetime(m.time_created\u002F1000, 'unixepoch', 'localtime') AS time,\n    substr(json_extract(p.data, '$.text'), 1, 200) AS snippet\n  FROM part p\n  JOIN message m ON m.id = p.message_id\n  JOIN session s ON s.id = m.session_id\n  WHERE s.parent_id IS NULL\n    AND json_extract(p.data, '$.type') = 'text'\n    AND json_extract(p.data, '$.text') LIKE '%SEARCH_TERM%'\n  ORDER BY m.time_created DESC\n  LIMIT 10;\n\"\n```\n\n### List saved plans\n\n```bash\nls -lt \"$DATA_ROOT\"\u002Fplans\u002F*.md 2>\u002Fdev\u002Fnull | head -20\n```\n\nTo read a specific plan:\n\n```bash\ncat \"$DATA_ROOT\"\u002Fplans\u002FFILENAME.md\n```\n\n### Show recent prompt history\n\n```bash\ntail -20 \"$STATE_ROOT\"\u002Fprompt-history.jsonl\n```\n\nEach line is a JSON object. The user's input is typically in the `input` or `text` field.\n\n## Workflow\n\n### Quick recall (most common)\n\n1. Run the **summary** query to see what's available.\n2. If you need sessions for the current project, get the worktree with `git rev-parse --show-toplevel`, then run the **project sessions** query.\n3. If you need a specific topic, run the **search** query.\n4. If you need full conversation detail, run the **messages** query with the session ID.\n\n### Plan review\n\n1. List plans with `ls -lt \"$DATA_ROOT\"\u002Fplans\u002F*.md`.\n2. Read a plan with `cat \"$DATA_ROOT\"\u002Fplans\u002F\u003Cfilename>.md`.\n\n### Deep investigation\n\n1. Run **projects** to see all tracked repos.\n2. Run **sessions** for a specific project.\n3. Run **messages** for full conversation content.\n4. Cross-reference with **search** across all projects.\n\n## Critical Rules\n\n1. **Read-only.** Never write to or modify the database or any OpenCode files.\n2. **Use bash + sqlite3.** Do not try to read `opencode.db` with the Read tool — it is a binary file. Always query via `sqlite3` in bash.\n3. **Don't dump everything.** Use `LIMIT` and `LIKE` to keep output focused. The database can contain tens of thousands of messages.\n4. **Summarize for the user.** After retrieving data, distill the relevant parts. Don't paste raw query output.\n5. **Respect privacy.** Session history may contain sensitive data. Only surface what is relevant to the current task.\n6. **Set path variables first.** At the start of any memory lookup, set `DATA_ROOT`, `STATE_ROOT`, `DB`, and `DB_URI` exactly as shown above so the commands work on XDG and non-XDG setups and keep SQLite access read-only.\n\n## Fallback: Web UI\n\nIf the user needs visual dashboards or a browsable interface:\n\n1. Check if OpenCode web is running: `curl -s http:\u002F\u002F127.0.0.1:4096\u002Fapi\u002Fhealth 2>\u002Fdev\u002Fnull || echo \"not running\"`\n2. If running, direct the user to `http:\u002F\u002F127.0.0.1:4096`.\n3. If not running, suggest `opencode web`.\n4. Note: `opencode.local` only works with mDNS enabled (`opencode web --mdns`). Don't assume it exists.\n\n## Deep Reference\n\nSee [references\u002Fstorage-format.md](references\u002Fstorage-format.md) for the full storage layout, all table schemas, and additional query examples.\n",{"name":5,"license":15,"description":46,"compatibility":47},"Browse and recall OpenCode local memory stored on the user's machine: local sessions, plans, conversations, prompt history, and project context. Use immediately when the user asks to check history, previous sessions, past chats, what did we do before, last time, check plans, session history, recall, memory, remember, prior work, previous context, or have we done this before. Auto-trigger proactively when resuming work, continuing a project, referencing prior decisions, debugging repeated issues, revisiting earlier plans, or any follow-up where earlier OpenCode context may help. This means OpenCode local history\u002Ffiles specifically, not ChatGPT\u002FClaude cloud history, generic web search, or unrelated product memory systems. Do NOT use for fresh tasks with no relevant history, or when current files\u002Fgit already answer the question.\n","opencode","master","8aa8e87ab2fed38a42acefb4a2d9a3a557f0b1ed","config\u002Fopencode\u002Fskills\u002Fopencode-memory",7,"https:\u002F\u002Fgithub.com\u002Fajoslin\u002Fdot","https:\u002F\u002Fgithub.com\u002Fajoslin","http:\u002F\u002Fajoslin.com","Full-stack JS","2026-07-04 04:23:52",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add ajoslin\u002Fdot --skill opencode-memory",{"tool":63,"command":64},"bunx","bunx skills add ajoslin\u002Fdot --skill opencode-memory",{"tool":66,"command":67},"pnpm","pnpm dlx skills add ajoslin\u002Fdot --skill opencode-memory","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fajoslin\u002Fdot\u002Fblob\u002Fmaster\u002Fconfig\u002Fopencode\u002Fskills\u002Fopencode-memory\u002FSKILL.md",true,[71],{"path":72,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fstorage-format.md","REFERENCE","参考文档",5565,"https:\u002F\u002Fgithub.com\u002Fajoslin\u002Fdot\u002Fblob\u002Fmaster\u002Fconfig\u002Fopencode\u002Fskills\u002Fopencode-memory\u002Freferences\u002Fstorage-format.md",[78,101],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":82,"licenseFileUrl":6,"sourceHtmlUrl":83,"categories":84,"keywords":88,"publishedAt":41},"01M6BVTZQAWBWJWAM6P71G51G6","vercel-composition-patterns","React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.","MIT","https:\u002F\u002Fgithub.com\u002Fajoslin\u002Fdot\u002Fblob\u002Fmaster\u002Fconfig\u002Fopencode\u002Fskills\u002Fcomposition-patterns\u002FSKILL.md",[85],{"categoryNo":86,"categoryName":87,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",[89,90,91,92,93,94,95,96,36,97,98,99,100],"vercel composition patterns","Composition","react","refactoring","api","Providers","component architecture","libraries","api changes","architecture","component","components",{"skillNo":102,"name":103,"nameZh":6,"description":104,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":82,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":112,"publishedAt":41},"01M6BVTTFM7JJ9DQABX9QB519R","architecture-diagram","Generate professional dark-themed system architecture diagrams as standalone HTML\u002FSVG files. Self-contained output with no external dependencies. Based on Cocoon AI's architecture-diagram-generator (MIT).","https:\u002F\u002Fgithub.com\u002Fajoslin\u002Fdot\u002Fblob\u002Fmaster\u002Fconfig\u002Fopencode\u002Fskills\u002Farchitecture-diagram\u002FSKILL.md",[107,108,111],{"categoryNo":86,"categoryName":87,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":109,"categoryName":110,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[113,114,98,115,30,116,117,118,119,120],"diagram","architecture diagram","excalidraw","visualization","generate","infrastructure","architecture diagrams","hello"]