{
  "markdown": "# Datarails Finance OS Plugin for Claude\n\nAnalyze financial data, detect anomalies, and build FP&A reports directly in Claude — connected live to your Datarails Finance OS.\n\n## Install\n\n### Claude Desktop (Cowork)\n\n1. Open Claude Desktop → **Browse plugins** → **Personal** tab\n2. Click **+** → **Add marketplace from GitHub**\n3. Enter `Datarails/dr-claude-code-plugins-re`\n4. Install **Datarails FinanceOS**\n\n> **If marketplace install fails** (a [known SSH issue](https://github.com/anthropics/claude-code/issues/26588)): download the ZIP from the [latest release](https://github.com/Datarails/dr-claude-code-plugins-re/releases/latest) and add it via **+** → **Upload plugin**.\n\n### Claude Code (Terminal)\n\n```\n/plugin marketplace add https://github.com/Datarails/dr-claude-code-plugins-re.git\n/plugin install datarails-financeos@datarails-marketplace\n```\n\n> Use the full HTTPS URL (not the `owner/repo` shorthand) — the shorthand triggers SSH cloning, which fails without SSH keys configured.\n\nManage it anytime: `/plugin` (view, enable/disable), `/plugin update`, `/plugin uninstall`. For local development, load a clone directly with `claude --plugin-dir ./dr-claude-code-plugins-re` (that session only).\n\n## Connect to Datarails\n\nThe Datarails connector is **bundled with the plugin** — there's no separate setup or `claude mcp add` step. The first time a skill needs data, a browser window opens for **OAuth login**; sign in once and you're connected. (In Claude Desktop you can also connect from **+** → **Connectors** → **Datarails**.)\n\n## Using the plugin\n\n**Just ask.** Describe what you want in plain language and Claude picks the right skill automatically:\n\n> \"Give me a financial summary for 2026.\"\n> \"What are our top expenses this year — flag anything unusual.\"\n> \"Build an FP&A intelligence workbook.\"\n\n**Or invoke a skill explicitly.** In Claude Code, type `/datarails-financeos:` and press **Tab** to autocomplete every command:\n\n```\n/datarails-financeos:tables                      # list & explore tables\n/datarails-financeos:financial-summary            # revenue / expense / margin snapshot\n/datarails-financeos:intelligence --year 2026     # full FP&A workbook\n```\n\n> The `datarails-financeos:` prefix is **required** — Claude Code namespaces every plugin skill to prevent collisions, so a bare `/tables` won't resolve.\n\n## Skills\n\nInvoke any of these as **`/datarails-financeos:<name>`**, or just describe the task and let Claude choose. Every skill discovers your org's tables, fields, scenarios, and account structure at runtime — nothing is hardcoded.\n\n### Explore & query\n| Skill | What it does |\n|-------|--------------|\n| `tables` | List tables, view schemas, inspect a field's values |\n| `profile` | Field statistics — null rates, cardinality, range outliers |\n| `query` | Fetch and filter records; value-list **and** advanced filters (ranges, comparisons, text, null checks, date ranges) |\n| `test` | Check which fields work as aggregation dimensions in your environment |\n\n### Snapshots & trends\n| Skill | What it does |\n|-------|--------------|\n| `financial-summary` | Period-scoped snapshot: revenue, expenses, gross profit & margin |\n| `revenue-trends` | Revenue growth rates, seasonality, composition over time |\n| `expense-analysis` | Top expense categories, concentration, monthly trend |\n\n### Reports & workbooks\n| Skill | What it does | Output |\n|-------|--------------|--------|\n| `intelligence` | Full-fiscal-year FP&A workbook with auto-detected insights (no PowerPoint) | Excel (up to 10 sheets) |\n| `insights` | Executive trend analysis and narrative | PowerPoint + Excel |\n| `dashboard` | Executive KPI dashboard | Excel + PowerPoint |\n| `departments` | Departmental P&L and comparison | Excel + PowerPoint |\n| `forecast-variance` | Multi-scenario variance — scenarios discovered at runtime (no fixed \"Budget\" assumed) | Excel + PowerPoint |\n| `extract` | Extract validated financial data (P&L, Balance Sheet, sourceable KPIs) | Excel |\n\n### Assurance & data quality\n| Skill | What it does | Output |\n|-------|--------------|--------|\n| `anomalies` | Detect data-quality anomalies in a table | inline |\n| `anomalies-report` | Full data-quality assessment with a quality score | Excel |\n| `reconciliation` | Independent-source consistency checks — cross-endpoint agreement, balance-sheet identity, roll-ups, scenario/period integrity. Validates the **data pipeline**, not source systems | Excel |\n| `audit` | Audit-support evidence package (completeness, reconciliation, mapping integrity, sampling). **Not** a SOX certification — access-control / change-management / ITGC evidence is out of scope | PDF + Excel |\n\n### Excel & drill-down\n| Skill | What it does |\n|-------|--------------|\n| `get-formula` | Generate Excel workbooks with live `DR.GET` formulas |\n| `drilldown` | Break a Datarails number into its line items — from a workbook cell, a pasted `DR.GET` formula, or a plain-language description (no file needed) |\n\n> A few short **command aliases** also resolve under the same namespace for common tasks — `/datarails-financeos:explore-tables`, `:data-check`, `:budget-comparison`, `:test-api` — each launches the matching skill above.\n\n## The intelligence workbook\n\nThe flagship report. Run it for a full FP&A book:\n\n```\n/datarails-financeos:intelligence --year 2026\n```\n\nIt generates **up to 10 sheets** — conditional sheets are included only when your data sources them, and omitted (never fabricated) otherwise:\n\n1. **Insights Dashboard** — top findings with severity\n2. **Expense Deep Dive** — top accounts, % of total\n3. **Variance Waterfall** — what changed and why\n4. **Trend Analysis** — 12-month trends\n5. **Anomaly Report** — auto-detected outliers\n6. **Vendor Analysis** — top vendors, concentration *(when vendor data exists)*\n7. **SaaS Metrics** — ARR, NRR, CAC, LTV *(when your data sources them)*\n8. **Sales Performance** — rep leaderboard *(when rep/bookings data exists)*\n9. **Cost Center P&L** — department detail\n10. **Raw Data** — pivot-ready\n\n## Authentication\n\nHandled automatically via **OAuth 2.0 + PKCE** when the connector is first used — a browser window opens for login, tokens refresh on their own, and there are no manual steps.\n\n## Troubleshooting\n\n| Problem | Fix |\n|---------|-----|\n| A command isn't found | Type `/datarails-financeos:` and press **Tab** — every command is under that prefix; a bare name won't resolve |\n| Tools not available | Connect the Datarails connector: **+** → **Connectors** → **Datarails** → **Connect** |\n| \"Not authenticated\" | Disconnect and reconnect the Datarails connector |\n| Skill picked the wrong table/field | Skills discover your financials table and fields automatically; if it guesses wrong, tell it which to use and it continues — there's no profile to configure |\n| Commands not showing | Restart Claude Desktop / Claude Code, and check the plugin is enabled (`/plugin`) |\n| Slow extraction | Row-by-row extraction pages 500 rows at a time and can take minutes on large tables; summary skills use aggregation, which returns in seconds |\n\nSee [SETUP.md](SETUP.md) for detailed setup and troubleshooting, or the [Getting Started Guide](docs/guides/GETTING_STARTED.md) for a ~15-minute hands-on walkthrough.\n\n## License\n\nMIT License — see LICENSE file.\n\n## Privacy\n\nThis plugin connects to the Datarails Finance OS service to analyze your financial data. See the [Datarails Privacy Policy](https://www.datarails.com/privacy-policy/) for how your data is collected, used, and protected.\n\n## Support\n\n- GitHub Issues: https://github.com/Datarails/dr-claude-code-plugins-re/issues\n- Datarails Support: support@datarails.com\n",
  "bytes": 7653,
  "sha": "173c6556f079edc8c2cb74c4f043f8daa70598987935b6bcd22a6851d5181165",
  "repo_slug": "datarails/dr-claude-code-plugins-re",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/plg_datarails_dr_claude_code_plugins_re_data_c49c0f65/readme"
}