[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CR31NXRXCBHXVCY2DXE7AV":3,"skill-same:01M7CR31NXRXCBHXVCY2DXE7AV":71},{"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":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":45,"commitSha":46,"skillDirectory":47,"forksCount":14,"repositoryHtmlUrl":48,"repositoryHomepageUrl":6,"ownerHtmlUrl":49,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":50,"install":51,"resources":64},"01M7CR31NXRXCBHXVCY2DXE7AV","excel-author",null,"Build auditable financial workbooks headless via openpyxl.","thalhath2234\u002Ftimely-app","thalhath2234","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F69391707?v=4","01M7CQNWTHSAP0D50RQY6W9YZQ","Thalhath","01M7CQNWVQN600VKQXRGTWCDPH",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Fhermes\u002Foptional-skills\u002Ffinance\u002Fexcel-author\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: excel-author\ndescription: Build auditable financial workbooks headless via openpyxl.\nversion: 1.0.0\nauthor: Anthropic (adapted by Nous Research)\nlicense: Apache-2.0\nplatforms: [linux, macos, windows]\nmetadata:\n  hermes:\n    tags: [excel, openpyxl, finance, spreadsheet, modeling]\n    related_skills: [microsoft-office, pptx-author, dcf-model, comps-analysis, lbo-model, 3-statement-model]\n---\n\n# excel-author\n\nProduce an .xlsx file on disk using `openpyxl`. Follow the banker-grade conventions below so the model is auditable, flexible, and reviewable by someone other than the person who built it.\n\nAdapted from Anthropic's `xlsx-author` and `audit-xls` skills in the [anthropics\u002Ffinancial-services](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Ffinancial-services) repo. The MCP \u002F Office-JS \u002F Cowork-specific branches of the originals are dropped — this skill assumes headless Python.\n\n## Output contract\n\n- Write to `.\u002Fout\u002F\u003Cname>.xlsx`. Create `.\u002Fout\u002F` if it does not exist.\n- Return the relative path in your final message so downstream tools can pick it up.\n- One logical model per file. Do not append to an existing workbook unless explicitly asked.\n\n## Setup\n\nUse `python` when it already imports `openpyxl` (the Hermes Docker image ships\nit). Otherwise, when `hermes` is on PATH, let Hermes build its shared, pinned\ndocuments environment under the Hermes home (safe to run from several agents\nat once); in a Docker sandbox without Hermes the block builds a venv in the\nsandbox home with the same pinned versions. It is the same block as the\nmicrosoft-office\u002Fpdf skills. Never `pip install` into the interpreter that\nruns Hermes:\n\n```bash\npython -c \"import openpyxl\" 2>\u002Fdev\u002Fnull && PY=python || if command -v hermes >\u002Fdev\u002Fnull 2>&1; then\n  PY=\"$(hermes pm environment documents --extra documents --extra documents-pdf)\"\nelse\n  PY=\"$HOME\u002F.cache\u002Fhermes-docs-venv\u002Fbin\u002Fpython\"\n  \"$PY\" -c \"import docx, openpyxl, pptx, reportlab, pypdf, pdfplumber, pypdfium2\" 2>\u002Fdev\u002Fnull || (\n    # Safe to run from several agents at once: each caller builds a private venv,\n    # a symlink publishes the first finished one atomically, the others delete theirs;\n    # a published build is replaced and deleted only while it still fails to import.\n    mkdir -p \"$HOME\u002F.cache\" && cd \"$HOME\u002F.cache\" && stage=\"$(mktemp -d hermes-docs-venv.XXXXXX)\" || exit 1\n    old=\"$(readlink hermes-docs-venv)\"  # the published build, if any (another agent may have just published it)\n    python3 -m venv \"$stage\" &&\n      \"$stage\u002Fbin\u002Fpython\" -m pip install -q --disable-pip-version-check python-docx==1.2.0 openpyxl==3.1.5 python-pptx==1.0.2 reportlab==5.1.0 pypdf==6.21.4 pdfplumber==0.11.10 pypdfium2==5.14.0 &&\n      \"$stage\u002Fbin\u002Fpython\" -c \"import docx, openpyxl, pptx, reportlab, pypdf, pdfplumber, pypdfium2\" &&\n      if [ -z \"$old\" ]; then python3 -c 'import os, sys; os.symlink(*sys.argv[1:])' \"$stage\" hermes-docs-venv\n      elif [ \"$(readlink hermes-docs-venv)\" = \"$old\" ] && ! \"$old\u002Fbin\u002Fpython\" -c \"import docx, openpyxl, pptx, reportlab, pypdf, pdfplumber, pypdfium2\" 2>\u002Fdev\u002Fnull; then\n        ln -s \"$stage\" \"$stage.lnk\" && python3 -c 'import os, sys; os.replace(*sys.argv[1:])' \"$stage.lnk\" hermes-docs-venv &&\n          case \"$old\" in hermes-docs-venv.??????) rm -rf \"$old\" ;; esac\n      fi 2>\u002Fdev\u002Fnull\n    [ \"$(readlink hermes-docs-venv)\" = \"$stage\" ] || rm -rf \"$stage\" \"$stage.lnk\"\n  )\nfi\nexport PY\n```\n\nRun the model-building Python with `\"$PY\"`.\n\n## Core conventions (non-negotiable)\n\n### Blue \u002F black \u002F green cell color\n- **Blue** (`Font(color=\"0000FF\")`) — hardcoded input a human entered. Revenue drivers, WACC inputs, terminal growth, market data.\n- **Black** (default) — formula. Every derived cell is a live Excel formula.\n- **Green** (`Font(color=\"006100\")`) — link to another sheet or external file.\n\nA reviewer can then scan the sheet and immediately see what's an assumption vs. what's computed.\n\n### Formulas over hardcodes\nEvery calculation cell MUST be a formula string, never a number computed in Python and pasted as a value.\n\n```python\n# WRONG — silent bug waiting to happen\nws[\"D20\"] = revenue_prior_year * (1 + growth)\n\n# CORRECT — flexes when the user changes the assumption\nws[\"D20\"] = \"=D19*(1+$B$8)\"\n```\n\nThe only hardcoded numbers permitted:\n1. Raw historical inputs (actual revenues, reported EBITDA, etc.)\n2. Assumption drivers the user is meant to flex (growth rates, WACC inputs, terminal g)\n3. Current market data (share price, debt balance) — with a cell comment documenting source + date\n\nIf you catch yourself computing a value in Python and writing the result, stop.\n\n### Named ranges for cross-sheet references\nUse named ranges for any figure referenced from another sheet, a deck, or a memo.\n\n```python\nfrom openpyxl.workbook.defined_name import DefinedName\nwb.defined_names[\"WACC\"] = DefinedName(\"WACC\", attr_text=\"Inputs!$C$8\")\n# then elsewhere:\ncalc[\"D30\"] = \"=D29\u002FWACC\"\n```\n\n### Balance checks tab\nInclude a `Checks` tab that ties everything and surfaces TRUE\u002FFALSE:\n- Balance sheet balances (assets = liabilities + equity)\n- Cash flow ties to period-over-period cash change on the BS\n- Sum-of-parts ties to consolidated totals\n- No rogue hardcodes inside calc ranges\n\nExample:\n```python\nchecks = wb.create_sheet(\"Checks\")\nchecks[\"A2\"] = \"BS balances\"\nchecks[\"B2\"] = \"=IS!D20-IS!D21-IS!D22\"\nchecks[\"C2\"] = \"=ABS(B2)\u003C0.01\"  # TRUE\u002FFALSE\n```\n\n### Cell comments on every hardcoded input\nAdd the comment AS you create the cell, not later.\n\n```python\nfrom openpyxl.comments import Comment\nws[\"C2\"] = 1_250_000_000\nws[\"C2\"].font = Font(color=\"0000FF\")\nws[\"C2\"].comment = Comment(\"Source: 10-K FY2024, p.47, revenue line\", \"analyst\")\n```\n\nFormat: `Source: [System\u002FDocument], [Date], [Reference], [URL if applicable]`.\n\nNever defer sourcing. Never write `TODO: add source`.\n\n## Skeleton: typical financial model\n\n```python\nfrom openpyxl import Workbook\nfrom openpyxl.styles import Font, PatternFill, Alignment, Border, Side\nfrom openpyxl.comments import Comment\nfrom openpyxl.utils import get_column_letter\nfrom pathlib import Path\n\nBLUE = Font(color=\"0000FF\")\nBLACK = Font(color=\"000000\")\nGREEN = Font(color=\"006100\")\nBOLD = Font(bold=True)\nHEADER_FILL = PatternFill(\"solid\", fgColor=\"1F4E79\")\nHEADER_FONT = Font(color=\"FFFFFF\", bold=True)\n\nwb = Workbook()\n\n# --- Inputs tab ---\ninp = wb.active\ninp.title = \"Inputs\"\ninp[\"A1\"] = \"MARKET DATA & KEY INPUTS\"\ninp[\"A1\"].font = HEADER_FONT\ninp[\"A1\"].fill = HEADER_FILL\ninp.merge_cells(\"A1:C1\")\n\ninp[\"B3\"] = \"Revenue FY2024\"\ninp[\"C3\"] = 1_250_000_000\ninp[\"C3\"].font = BLUE\ninp[\"C3\"].comment = Comment(\"Source: 10-K FY2024 p.47\", \"model\")\n\ninp[\"B4\"] = \"Growth Rate\"\ninp[\"C4\"] = 0.12\ninp[\"C4\"].font = BLUE\n\n# --- Calc tab ---\ncalc = wb.create_sheet(\"DCF\")\ncalc[\"B2\"] = \"Projected Revenue\"\ncalc[\"C2\"] = \"=Inputs!C3*(1+Inputs!C4)\"   # formula, black\n\n# --- Checks tab ---\nchk = wb.create_sheet(\"Checks\")\nchk[\"A2\"] = \"BS balances\"\nchk[\"B2\"] = \"=ABS(BS!D20-BS!D21-BS!D22)\u003C0.01\"\n\nPath(\".\u002Fout\").mkdir(exist_ok=True)\nwb.save(\".\u002Fout\u002Fmodel.xlsx\")\n```\n\n## Section headers with merged cells\n\nopenpyxl quirk: when you merge, set the value on the top-left cell and style the full range separately.\n\n```python\nws[\"A7\"] = \"CASH FLOW PROJECTION\"\nws[\"A7\"].font = HEADER_FONT\nws.merge_cells(\"A7:H7\")\nfor col in range(1, 9):  # A..H\n    ws.cell(row=7, column=col).fill = HEADER_FILL\n```\n\n## Sensitivity tables\n\nBuild with loops, not hardcoded formulas per cell. Rules:\n\n- **Odd number of rows\u002Fcols** (5×5 or 7×7) — guarantees a true center cell.\n- **Center cell = base case.** The middle row\u002Fcol header must equal the model's actual WACC and terminal g so the center output equals the base-case implied share price. That's the sanity check.\n- **Highlight the center cell** with medium-blue fill (`\"BDD7EE\"`) and bold.\n- Populate every cell with a full recalculation formula — never an approximation.\n\n```python\n# 5x5 WACC (rows) x terminal growth (cols) sensitivity\nwacc_axis = [0.08, 0.085, 0.09, 0.095, 0.10]        # center row = base 9.0%\nterm_axis = [0.02, 0.025, 0.03, 0.035, 0.04]        # center col = base 3.0%\n\nstart_row = 40\nws.cell(row=start_row, column=1).value = \"Implied Share Price ($)\"\nws.cell(row=start_row, column=1).font = BOLD\n\nfor j, g in enumerate(term_axis):\n    ws.cell(row=start_row+1, column=2+j).value = g\n    ws.cell(row=start_row+1, column=2+j).font = BLUE\n\nfor i, w in enumerate(wacc_axis):\n    r = start_row + 2 + i\n    ws.cell(row=r, column=1).value = w\n    ws.cell(row=r, column=1).font = BLUE\n    for j, g in enumerate(term_axis):\n        c = 2 + j\n        # Full DCF recalc formula (simplified for illustration).\n        # In a real model this references the full projection block.\n        ws.cell(row=r, column=c).value = (\n            f\"=SUMPRODUCT(FCF_range,1\u002F(1+{w})^year_offset) + \"\n            f\"FCF_terminal*(1+{g})\u002F({w}-{g})\u002F(1+{w})^terminal_year\"\n        )\n\n# Highlight center cell (base case)\ncenter = ws.cell(row=start_row+2+len(wacc_axis)\u002F\u002F2,\n                 column=2+len(term_axis)\u002F\u002F2)\ncenter.fill = PatternFill(\"solid\", fgColor=\"BDD7EE\")\ncenter.font = BOLD\n```\n\n## Recalculating before delivery\n\nopenpyxl writes formula strings but does not compute them. Excel recalculates on open, but downstream consumers (auto-check scripts, CI) need computed values.\n\nRun LibreOffice or a dedicated recalc step before delivery:\n\n```bash\n# LibreOffice headless recalc\nlibreoffice --headless --calc --convert-to xlsx .\u002Fout\u002Fmodel.xlsx --outdir .\u002Fout\u002F\n```\n\nOr use a Python recalc helper (see `scripts\u002Frecalc.py` in this skill).\n\n## Model layout planning\n\nBefore writing any formula:\n1. Define ALL section row positions\n2. Write ALL headers and labels\n3. Write ALL section dividers and blank rows\n4. THEN write formulas using the locked row positions\n\nThis prevents the cascading-formula-breakage pattern where inserting a header row after formulas are written shifts every downstream reference.\n\n## Verify step-by-step with the user\n\nFor large models (DCFs, 3-statement, LBO), stop and show the user intermediate artifacts before continuing. Catching a wrong margin assumption before you've built downstream sensitivity tables saves an hour.\n\nCheckpoint pattern:\n- After Inputs block → show raw inputs, confirm before projecting\n- After Revenue projections → confirm top line + growth\n- After FCF build → confirm the full schedule\n- After WACC → confirm inputs\n- After valuation → confirm the equity bridge\n- THEN build sensitivity tables\n\n## When NOT to use this skill\n\n- Users in a live Excel session with an Office MCP available — drive their live workbook instead.\n- Pure tabular data export with no formulas — `csv` or `pandas.to_excel` is simpler.\n- Dashboards \u002F charts with heavy interactivity — use a real BI tool.\n\n## Attribution\n\nConventions (blue\u002Fblack\u002Fgreen, formulas-over-hardcodes, named ranges, sensitivity rules) adapted from Anthropic's Claude for Financial Services plugin suite, Apache-2.0 licensed. Original: https:\u002F\u002Fgithub.com\u002Fanthropics\u002Ffinancial-services\u002Ftree\u002Fmain\u002Fplugins\u002Fvertical-plugins\u002Ffinancial-analysis\u002Fskills\u002Fxlsx-author\n","# excel-author\n\nProduce an .xlsx file on disk using `openpyxl`. Follow the banker-grade conventions below so the model is auditable, flexible, and reviewable by someone other than the person who built it.\n\nAdapted from Anthropic's `xlsx-author` and `audit-xls` skills in the [anthropics\u002Ffinancial-services](https:\u002F\u002Fgithub.com\u002Fanthropics\u002Ffinancial-services) repo. The MCP \u002F Office-JS \u002F Cowork-specific branches of the originals are dropped — this skill assumes headless Python.\n\n## Output contract\n\n- Write to `.\u002Fout\u002F\u003Cname>.xlsx`. Create `.\u002Fout\u002F` if it does not exist.\n- Return the relative path in your final message so downstream tools can pick it up.\n- One logical model per file. Do not append to an existing workbook unless explicitly asked.\n\n## Setup\n\nUse `python` when it already imports `openpyxl` (the Hermes Docker image ships\nit). Otherwise, when `hermes` is on PATH, let Hermes build its shared, pinned\ndocuments environment under the Hermes home (safe to run from several agents\nat once); in a Docker sandbox without Hermes the block builds a venv in the\nsandbox home with the same pinned versions. It is the same block as the\nmicrosoft-office\u002Fpdf skills. Never `pip install` into the interpreter that\nruns Hermes:\n\n```bash\npython -c \"import openpyxl\" 2>\u002Fdev\u002Fnull && PY=python || if command -v hermes >\u002Fdev\u002Fnull 2>&1; then\n  PY=\"$(hermes pm environment documents --extra documents --extra documents-pdf)\"\nelse\n  PY=\"$HOME\u002F.cache\u002Fhermes-docs-venv\u002Fbin\u002Fpython\"\n  \"$PY\" -c \"import docx, openpyxl, pptx, reportlab, pypdf, pdfplumber, pypdfium2\" 2>\u002Fdev\u002Fnull || (\n    # Safe to run from several agents at once: each caller builds a private venv,\n    # a symlink publishes the first finished one atomically, the others delete theirs;\n    # a published build is replaced and deleted only while it still fails to import.\n    mkdir -p \"$HOME\u002F.cache\" && cd \"$HOME\u002F.cache\" && stage=\"$(mktemp -d hermes-docs-venv.XXXXXX)\" || exit 1\n    old=\"$(readlink hermes-docs-venv)\"  # the published build, if any (another agent may have just published it)\n    python3 -m venv \"$stage\" &&\n      \"$stage\u002Fbin\u002Fpython\" -m pip install -q --disable-pip-version-check python-docx==1.2.0 openpyxl==3.1.5 python-pptx==1.0.2 reportlab==5.1.0 pypdf==6.21.4 pdfplumber==0.11.10 pypdfium2==5.14.0 &&\n      \"$stage\u002Fbin\u002Fpython\" -c \"import docx, openpyxl, pptx, reportlab, pypdf, pdfplumber, pypdfium2\" &&\n      if [ -z \"$old\" ]; then python3 -c 'import os, sys; os.symlink(*sys.argv[1:])' \"$stage\" hermes-docs-venv\n      elif [ \"$(readlink hermes-docs-venv)\" = \"$old\" ] && ! \"$old\u002Fbin\u002Fpython\" -c \"import docx, openpyxl, pptx, reportlab, pypdf, pdfplumber, pypdfium2\" 2>\u002Fdev\u002Fnull; then\n        ln -s \"$stage\" \"$stage.lnk\" && python3 -c 'import os, sys; os.replace(*sys.argv[1:])' \"$stage.lnk\" hermes-docs-venv &&\n          case \"$old\" in hermes-docs-venv.??????) rm -rf \"$old\" ;; esac\n      fi 2>\u002Fdev\u002Fnull\n    [ \"$(readlink hermes-docs-venv)\" = \"$stage\" ] || rm -rf \"$stage\" \"$stage.lnk\"\n  )\nfi\nexport PY\n```\n\nRun the model-building Python with `\"$PY\"`.\n\n## Core conventions (non-negotiable)\n\n### Blue \u002F black \u002F green cell color\n- **Blue** (`Font(color=\"0000FF\")`) — hardcoded input a human entered. Revenue drivers, WACC inputs, terminal growth, market data.\n- **Black** (default) — formula. Every derived cell is a live Excel formula.\n- **Green** (`Font(color=\"006100\")`) — link to another sheet or external file.\n\nA reviewer can then scan the sheet and immediately see what's an assumption vs. what's computed.\n\n### Formulas over hardcodes\nEvery calculation cell MUST be a formula string, never a number computed in Python and pasted as a value.\n\n```python\n# WRONG — silent bug waiting to happen\nws[\"D20\"] = revenue_prior_year * (1 + growth)\n\n# CORRECT — flexes when the user changes the assumption\nws[\"D20\"] = \"=D19*(1+$B$8)\"\n```\n\nThe only hardcoded numbers permitted:\n1. Raw historical inputs (actual revenues, reported EBITDA, etc.)\n2. Assumption drivers the user is meant to flex (growth rates, WACC inputs, terminal g)\n3. Current market data (share price, debt balance) — with a cell comment documenting source + date\n\nIf you catch yourself computing a value in Python and writing the result, stop.\n\n### Named ranges for cross-sheet references\nUse named ranges for any figure referenced from another sheet, a deck, or a memo.\n\n```python\nfrom openpyxl.workbook.defined_name import DefinedName\nwb.defined_names[\"WACC\"] = DefinedName(\"WACC\", attr_text=\"Inputs!$C$8\")\n# then elsewhere:\ncalc[\"D30\"] = \"=D29\u002FWACC\"\n```\n\n### Balance checks tab\nInclude a `Checks` tab that ties everything and surfaces TRUE\u002FFALSE:\n- Balance sheet balances (assets = liabilities + equity)\n- Cash flow ties to period-over-period cash change on the BS\n- Sum-of-parts ties to consolidated totals\n- No rogue hardcodes inside calc ranges\n\nExample:\n```python\nchecks = wb.create_sheet(\"Checks\")\nchecks[\"A2\"] = \"BS balances\"\nchecks[\"B2\"] = \"=IS!D20-IS!D21-IS!D22\"\nchecks[\"C2\"] = \"=ABS(B2)\u003C0.01\"  # TRUE\u002FFALSE\n```\n\n### Cell comments on every hardcoded input\nAdd the comment AS you create the cell, not later.\n\n```python\nfrom openpyxl.comments import Comment\nws[\"C2\"] = 1_250_000_000\nws[\"C2\"].font = Font(color=\"0000FF\")\nws[\"C2\"].comment = Comment(\"Source: 10-K FY2024, p.47, revenue line\", \"analyst\")\n```\n\nFormat: `Source: [System\u002FDocument], [Date], [Reference], [URL if applicable]`.\n\nNever defer sourcing. Never write `TODO: add source`.\n\n## Skeleton: typical financial model\n\n```python\nfrom openpyxl import Workbook\nfrom openpyxl.styles import Font, PatternFill, Alignment, Border, Side\nfrom openpyxl.comments import Comment\nfrom openpyxl.utils import get_column_letter\nfrom pathlib import Path\n\nBLUE = Font(color=\"0000FF\")\nBLACK = Font(color=\"000000\")\nGREEN = Font(color=\"006100\")\nBOLD = Font(bold=True)\nHEADER_FILL = PatternFill(\"solid\", fgColor=\"1F4E79\")\nHEADER_FONT = Font(color=\"FFFFFF\", bold=True)\n\nwb = Workbook()\n\n# --- Inputs tab ---\ninp = wb.active\ninp.title = \"Inputs\"\ninp[\"A1\"] = \"MARKET DATA & KEY INPUTS\"\ninp[\"A1\"].font = HEADER_FONT\ninp[\"A1\"].fill = HEADER_FILL\ninp.merge_cells(\"A1:C1\")\n\ninp[\"B3\"] = \"Revenue FY2024\"\ninp[\"C3\"] = 1_250_000_000\ninp[\"C3\"].font = BLUE\ninp[\"C3\"].comment = Comment(\"Source: 10-K FY2024 p.47\", \"model\")\n\ninp[\"B4\"] = \"Growth Rate\"\ninp[\"C4\"] = 0.12\ninp[\"C4\"].font = BLUE\n\n# --- Calc tab ---\ncalc = wb.create_sheet(\"DCF\")\ncalc[\"B2\"] = \"Projected Revenue\"\ncalc[\"C2\"] = \"=Inputs!C3*(1+Inputs!C4)\"   # formula, black\n\n# --- Checks tab ---\nchk = wb.create_sheet(\"Checks\")\nchk[\"A2\"] = \"BS balances\"\nchk[\"B2\"] = \"=ABS(BS!D20-BS!D21-BS!D22)\u003C0.01\"\n\nPath(\".\u002Fout\").mkdir(exist_ok=True)\nwb.save(\".\u002Fout\u002Fmodel.xlsx\")\n```\n\n## Section headers with merged cells\n\nopenpyxl quirk: when you merge, set the value on the top-left cell and style the full range separately.\n\n```python\nws[\"A7\"] = \"CASH FLOW PROJECTION\"\nws[\"A7\"].font = HEADER_FONT\nws.merge_cells(\"A7:H7\")\nfor col in range(1, 9):  # A..H\n    ws.cell(row=7, column=col).fill = HEADER_FILL\n```\n\n## Sensitivity tables\n\nBuild with loops, not hardcoded formulas per cell. Rules:\n\n- **Odd number of rows\u002Fcols** (5×5 or 7×7) — guarantees a true center cell.\n- **Center cell = base case.** The middle row\u002Fcol header must equal the model's actual WACC and terminal g so the center output equals the base-case implied share price. That's the sanity check.\n- **Highlight the center cell** with medium-blue fill (`\"BDD7EE\"`) and bold.\n- Populate every cell with a full recalculation formula — never an approximation.\n\n```python\n# 5x5 WACC (rows) x terminal growth (cols) sensitivity\nwacc_axis = [0.08, 0.085, 0.09, 0.095, 0.10]        # center row = base 9.0%\nterm_axis = [0.02, 0.025, 0.03, 0.035, 0.04]        # center col = base 3.0%\n\nstart_row = 40\nws.cell(row=start_row, column=1).value = \"Implied Share Price ($)\"\nws.cell(row=start_row, column=1).font = BOLD\n\nfor j, g in enumerate(term_axis):\n    ws.cell(row=start_row+1, column=2+j).value = g\n    ws.cell(row=start_row+1, column=2+j).font = BLUE\n\nfor i, w in enumerate(wacc_axis):\n    r = start_row + 2 + i\n    ws.cell(row=r, column=1).value = w\n    ws.cell(row=r, column=1).font = BLUE\n    for j, g in enumerate(term_axis):\n        c = 2 + j\n        # Full DCF recalc formula (simplified for illustration).\n        # In a real model this references the full projection block.\n        ws.cell(row=r, column=c).value = (\n            f\"=SUMPRODUCT(FCF_range,1\u002F(1+{w})^year_offset) + \"\n            f\"FCF_terminal*(1+{g})\u002F({w}-{g})\u002F(1+{w})^terminal_year\"\n        )\n\n# Highlight center cell (base case)\ncenter = ws.cell(row=start_row+2+len(wacc_axis)\u002F\u002F2,\n                 column=2+len(term_axis)\u002F\u002F2)\ncenter.fill = PatternFill(\"solid\", fgColor=\"BDD7EE\")\ncenter.font = BOLD\n```\n\n## Recalculating before delivery\n\nopenpyxl writes formula strings but does not compute them. Excel recalculates on open, but downstream consumers (auto-check scripts, CI) need computed values.\n\nRun LibreOffice or a dedicated recalc step before delivery:\n\n```bash\n# LibreOffice headless recalc\nlibreoffice --headless --calc --convert-to xlsx .\u002Fout\u002Fmodel.xlsx --outdir .\u002Fout\u002F\n```\n\nOr use a Python recalc helper (see `scripts\u002Frecalc.py` in this skill).\n\n## Model layout planning\n\nBefore writing any formula:\n1. Define ALL section row positions\n2. Write ALL headers and labels\n3. Write ALL section dividers and blank rows\n4. THEN write formulas using the locked row positions\n\nThis prevents the cascading-formula-breakage pattern where inserting a header row after formulas are written shifts every downstream reference.\n\n## Verify step-by-step with the user\n\nFor large models (DCFs, 3-statement, LBO), stop and show the user intermediate artifacts before continuing. Catching a wrong margin assumption before you've built downstream sensitivity tables saves an hour.\n\nCheckpoint pattern:\n- After Inputs block → show raw inputs, confirm before projecting\n- After Revenue projections → confirm top line + growth\n- After FCF build → confirm the full schedule\n- After WACC → confirm inputs\n- After valuation → confirm the equity bridge\n- THEN build sensitivity tables\n\n## When NOT to use this skill\n\n- Users in a live Excel session with an Office MCP available — drive their live workbook instead.\n- Pure tabular data export with no formulas — `csv` or `pandas.to_excel` is simpler.\n- Dashboards \u002F charts with heavy interactivity — use a real BI tool.\n\n## Attribution\n\nConventions (blue\u002Fblack\u002Fgreen, formulas-over-hardcodes, named ranges, sensitivity rules) adapted from Anthropic's Claude for Financial Services plugin suite, Apache-2.0 licensed. Original: https:\u002F\u002Fgithub.com\u002Fanthropics\u002Ffinancial-services\u002Ftree\u002Fmain\u002Fplugins\u002Fvertical-plugins\u002Ffinancial-analysis\u002Fskills\u002Fxlsx-author\n",{"name":5,"author":24,"license":15,"version":25,"metadata":26,"platforms":41,"description":7},"Anthropic (adapted by Nous Research)","1.0.0",{"hermes":27},{"tags":28,"related_skills":34},[29,30,31,32,33],"excel","openpyxl","finance","spreadsheet","modeling",[35,36,37,38,39,40],"microsoft-office","pptx-author","dcf-model","comps-analysis","lbo-model","3-statement-model",[42,43,44],"linux","macos","windows","dev","4d9b346961dd1d220a07fe4e6ed44b52daa57e0b","apps\u002Fagent\u002Fhermes\u002Foptional-skills\u002Ffinance\u002Fexcel-author","https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app","https:\u002F\u002Fgithub.com\u002Fthalhath2234","2026-11-14 05:26:10",{"source":8,"skillName":5,"commands":52,"prompt":62,"zipAvailable":63},[53,56,59],{"tool":54,"command":55},"npx","npx skills add thalhath2234\u002Ftimely-app --skill excel-author",{"tool":57,"command":58},"bunx","bunx skills add thalhath2234\u002Ftimely-app --skill excel-author",{"tool":60,"command":61},"pnpm","pnpm dlx skills add thalhath2234\u002Ftimely-app --skill excel-author","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Fhermes\u002Foptional-skills\u002Ffinance\u002Fexcel-author\u002FSKILL.md",true,[65],{"path":66,"resourceType":67,"resourceTypeDesc":68,"sizeBytes":69,"sourceHtmlUrl":70},"scripts\u002Frecalc.py","SCRIPT","脚本",2750,"https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Fhermes\u002Foptional-skills\u002Ffinance\u002Fexcel-author\u002Fscripts\u002Frecalc.py",[72,80,87,94,101,108],{"skillNo":73,"name":74,"nameZh":6,"description":75,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":6,"sourceHtmlUrl":77,"categories":78,"keywords":79,"publishedAt":19},"01M7CRJRHHG68Q3XZX37TD85JB","timely","Test fixture standing in for the Timely skill (IDs, markdown dialect, workspace rule).","MIT","https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Ftests\u002Ffixtures\u002Fskills\u002Ftimely\u002FSKILL.md",[],[],{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7CRJPPNS8SKWVTDP3TJ9EHG","product-price-monitor","Test fixture override of the upstream product-price-monitor skill.","https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Ftests\u002Ffixtures\u002Fskills\u002Foverrides\u002Fproduct-price-monitor\u002FSKILL.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M7CRJKRY4QH4RAA2MKHNZT04","blocked-page-recovery","Use when a fetch fails: 403\u002F429, paywall, WAF, bot wall.","https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Fhermes\u002Fskills\u002Fweb\u002Fblocked-page-recovery\u002FSKILL.md",[],[],{"skillNo":95,"name":96,"nameZh":6,"description":97,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":6,"sourceHtmlUrl":98,"categories":99,"keywords":100,"publishedAt":19},"01M7CRJG3JC6PRSTH3MW62KTH6","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Fhermes\u002Fskills\u002Fsoftware-development\u002Ftest-driven-development\u002FSKILL.md",[],[],{"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":76,"licenseFileUrl":6,"sourceHtmlUrl":105,"categories":106,"keywords":107,"publishedAt":19},"01M7CRJ8K3ACJYKSYDJWKJM6HV","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Fhermes\u002Fskills\u002Fsoftware-development\u002Fsystematic-debugging\u002FSKILL.md",[],[],{"skillNo":109,"name":110,"nameZh":6,"description":111,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":76,"licenseFileUrl":6,"sourceHtmlUrl":112,"categories":113,"keywords":114,"publishedAt":19},"01M7CRJ3Q5QP1CVBG959M3A3BM","spike","Throwaway experiments to validate an idea before build.","https:\u002F\u002Fgithub.com\u002Fthalhath2234\u002Ftimely-app\u002Fblob\u002Fdev\u002Fapps\u002Fagent\u002Fhermes\u002Fskills\u002Fsoftware-development\u002Fspike\u002FSKILL.md",[],[]]