ads-audit
Full multi-platform paid advertising audit with parallel subagent delegation. Analyzes Google Ads, Meta Ads, LinkedIn Ads, TikTok…
Set up an agentic OS inside an Obsidian vault — a configurable command-center dashboard with 5 auto-installed, bundled plugins (Dataview, CustomJS, Shell-commands, Terminal, Homepage), Home + per-profile + Vault Overview pages, KPI cards, sparklines, heatmap, task rollup, and a
Claude Code
$ npx -y skills add naveedharri/benai-skills --skill agentic-os-obsidian --agent claude-code
Codex
$ npx skills add naveedharri/benai-skills --skill agentic-os-obsidian -a codex -y
Cursor
$ npx skills add naveedharri/benai-skills --skill agentic-os-obsidian -a cursor -y
Gemini CLI
$ npx skills add naveedharri/benai-skills --skill agentic-os-obsidian -a gemini-cli -y
GitHub Copilot
$ npx skills add naveedharri/benai-skills --skill agentic-os-obsidian -a github-copilot -y
OpenCode
$ npx skills add naveedharri/benai-skills --skill agentic-os-obsidian -a opencode -y
Windsurf
$ npx skills add naveedharri/benai-skills --skill agentic-os-obsidian -a windsurf -y
name: agentic-os-obsidian description: Set up an agentic OS inside an Obsidian vault — a configurable command-center dashboard with 5 auto-installed, bundled plugins (Dataview, CustomJS, Shell-commands, Terminal, Homepage), Home + per-profile + Vault Overview pages, KPI cards, sparklines, heatmap, task rollup, and a button bar wired to user-defined Claude prompts. Markdown-native, no servers. Use when the user says "set up agentic OS in Obsidian", "install command center in my vault", "build a vault dashboard", "give me my own version of the dashboard inside Obsidian", "set up my second brain dashboard", or asks to personalize a vault dashboard previously created with this skill. disable-model-invocation: true allowed-tools: Bash, Read, Write, Edit, AskUserQuestion
Installs a configurable command-center dashboard *inside* an existing Obsidian vault. Bundles and side-loads 5 plugins (Dataview, CustomJS, Shell-commands, Terminal, Homepage). Home + per-profile + Vault Overview pages, KPI cards, sparklines, heatmap, task rollup, recent-files lists, and a button bar wired to user-defined Claude prompts via the Shell-commands plugin. All 5 required plugins are bundled and auto-enabled — the user only has to restart Obsidian. Markdown-based, no servers.
> Looking for a real web app with live API integrations and a Railway URL instead? That's the standalone route — use the `agentic-os-standalone` skill.
{vault}/
├── Dashboard/
│ ├── Home.md # Team-wide dashboard
│ ├── Vault-Overview.md # Top-level vault health view
│ ├── {Profile}.md # One per configured profile
│ ├── Setup.md # In-vault setup reference
│ ├── CLAUDE.md # AI routing rules for this folder
│ ├── components/dashboard.js # The renderer (CONFIG substituted)
│ └── lib/frappe-charts.min.js
└── .obsidian/
├── snippets/command-center.css # Brand-styled CSS (enabled)
├── plugins/dataview/ # Side-loaded, enabled
├── plugins/customjs/ # Side-loaded, enabled
├── plugins/obsidian-shellcommands/ # Side-loaded, enabled, aliases pre-registered
├── plugins/terminal/ # Side-loaded, enabled, "Claude" profile configured
├── plugins/homepage/ # Side-loaded, enabled, points to Dashboard/Home
├── community-plugins.json # Union'd with the 5 plugin IDs
└── appearance.json # CSS snippet enabledtest -d .obsidian && echo "vault confirmed" || (echo "NOT A VAULT — cd into your Obsidian vault first" && exit 1) test -d Dashboard && echo "DASHBOARD EXISTS — ask user before overwriting"
If `Dashboard/` exists, ask: cancel / install into `Dashboard-new/` / overwrite (require explicit "yes overwrite").
Use `AskUserQuestion` for each. Capture answers into variables for substitution into CONFIG.
"What's your organization or personal brand name? (Used in folder paths and dashboard header. Short — one word ideal.)" Default if skipped: `MyOrg`. Variable: `ORG_NAME`.
"What should the dashboard show as its title?" Two parts:
"Who's on the team? List each profile name on a separate line. Use lowercase or capitalized — must match the folder names you'll use in step 4. For a solo vault, just one name." Parse into array. Variable: `PROFILES` (JSON array). The first is `DEFAULT_PROFILE`.
"Where do you keep per-profile data?" Show 4 options via AskUserQuestion:
Variable: `PROFILE_FOLDER_PATTERN`. Verify each `PROFILES[i]` folder exists (warn if missing — installer will still write the dashboard; profile folders just won't have data yet).
Three quick AskUserQuestion calls (each with sensible defaults):
Also: **Root daily path** for team-wide rollups. Default `Daily`. Variable: `ROOT_DAILY_PATH`.
Which dashboard pages? MultiSelect:
Variable: `PAGES` (array). Installer creates one `.md` per page.
"Name 3-5 top-level folders to show on the Vault Overview page. For each: folder path, one-line description, optional CLAUDE.md path (or empty)." Variable: `OVERVIEW_FOLDERS` (array of `[path, description, claude_path]`). Defaults if skipped: detect `Projects/`, `Resources/`, `Intelligence/`, `Departments/`, `Teams/`, `Daily/` if present.
"What action buttons should the dashboard expose? For each button: label (short), icon (emoji or empty), scope (team / profile / both), and a Claude prompt (the text passed to `claude -p '...'`). Use `{profile}` in the prompt to interpolate the current profile name. Add as many as you want; defaults are zero."
Suggest these as star
Full multi-platform paid advertising audit with parallel subagent delegation. Analyzes Google Ads, Meta Ads, LinkedIn Ads, TikTok…
Budget allocation and bidding strategy review across all ad platforms. Evaluates spend distribution, bidding strategy…
Competitor ad intelligence analysis across Google, Meta, LinkedIn, TikTok, and Microsoft. Analyzes competitor ad copy, creative…
Cross-platform creative quality audit covering ad copy, video, image, and format diversity across all platforms. Detects creative…
Google Ads deep analysis covering Search, Performance Max, Display, YouTube, and Demand Gen campaigns. Evaluates 74 checks across…
Landing page quality assessment for paid advertising campaigns. Evaluates message match, page speed, mobile experience, trust…
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work. Also covers…
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling…