financials-normalizer — Codex skill: what it does, how to install
Standardizes public company financial filings into audit-ready tables. Best for analysts structuring messy disclosures before building financial models.
npx skills add openai/plugins --skill financials-normalizer -a codexWhat financials-normalizer does
The skill parses corporate earnings releases, market figures, and investor decks into uniform datasets. Output artifacts include Source_Index.csv, Normalized_Financials_Long.csv, and a defect log in Normalization_Issues.csv. For final reviews, the workflow exports tabular schedules in CSV or formatted XLSX workbooks.
Fabricating missing figures is strictly forbidden; unavailable entries get assigned a missing_required_source tag. Whenever a citation identifier is omitted, the record flags SRC-UNSPECIFIED and halts downstream progression. If accounting presentation changes or segments shift across periods, it constructs a comparability bridge instead of overwriting raw numbers.
Use this workflow strictly for public corporate statements. Feeding private deal-room documents or non-accounting data into it is explicitly ruled out.
What it helps with
- standardize messy public corporate disclosures into clean tables
- track citations, reporting currencies, and timestamps across periods
- build comparability bridges when issuers adjust segment reporting
- surface audit discrepancies and unaudited inputs before modeling
Example prompts for the agent
- Normalize the income statement and balance sheet lines from this 10-K into a workbook
- Extract public segment revenue figures and generate the source index log
- Audit historical reporting variances and output the normalization issues list
How to install financials-normalizer
1Skills CLI (Vercel)
npx skills add openai/plugins --skill financials-normalizer -a codex2Manually: put the files in place
.agents/skills/financials-normalizer/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add openai/plugins --skill financials-normalizer -a claude-code2Manually: put the files in place
~/.claude/skills/financials-normalizer/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/financials-normalizer/
Manually: put the files in place
.cursor/skills/financials-normalizer/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install openai/plugins financials-normalizer2Manually: put the files in place
.github/skills/financials-normalizer/SKILL.mdPlatform CLI
gemini skills install https://github.com/openai/plugins --path plugins/public-equity-investing/skills/financials-normalizerManually: put the files in place
.devin/skills/financials-normalizer/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: Codex skills: how to install, where they live, how to call them
Questions about financials-normalizer
How do I install financials-normalizer in Codex?
Run npx skills add openai/plugins --skill financials-normalizer -a codex in your terminal. This places the definition at .agents/skills/financials-normalizer/SKILL.md.
What does financials-normalizer do?
It transforms raw public company disclosures into normalized tables, tracking citations, period dates, and accounting discrepancies for modeling. It does not handle private data rooms.
How does it handle missing values or citations?
Empty values are marked as missing_required_source rather than guessed. Any record lacking an explicit source identifier receives an SRC-UNSPECIFIED flag, blocking readiness.
