A comprehensive collection of AI agent skills for financial professionals, accountants, and finance teams. Built for technical finance professionals who want AI coding agents to help with financial analysis, auditing, compliance, reporting, and financial engineering. Works with Claude Code, OpenAI Codex, Cursor, Windsurf, and any agent that supports the Agent Skills spec.
Built by Ayotunde Oso (aka GAJETOso) for accountants in oil & gas, banking, insurance, and corporate finance. Need specific help? Open an issue — this is a community-driven resource.
Contributions welcome! Found a way to improve a skill or have a new one to add? Open a PR.
Run into a problem or have a question? Open an issue — we're happy to help.
Skills are markdown files that give AI agents specialized knowledge and workflows for specific financial tasks. When you add these to your project, your agent can recognize when you're working on a financial task and apply the right accounting standards, compliance frameworks, and best practices.
Related MCP server: Jaz AI
Skills reference each other and build on shared financial context. Most skills check for foundational knowledge (accounting standards, regulatory requirements) before executing specialized tasks.
┌──────────────────────────────────────┐
│ Compliance & Standards │
│ (IFRS, GAAP, Local Regulations) │
└──────────────┬───────────────────────┘
│
┌──────────────┬────────────┬──────────┼──────────┬──────────────┬──────────────┐
▼ ▼ ▼ ▼ ▼ ▼ ▼
┌──────────┐ ┌──────────┐ ┌──────────┐ ┌──────┐ ┌──────────┐ ┌─────────────┐ ┌───────────┐
│ Financial│ │ Audit & │ │ Tax & │ │ Risk │ │Corporate │ │ Industry │ │ Advanced │
│ Analysis │ │Compliance│ │Regulatory│ │ │ │ Finance │ │ Specific │ │ Analytics │
├──────────┤ ├──────────┤ ├──────────┤ ├──────┤ ├──────────┤ ├─────────────┤ ├───────────┤
│fin-anlys │ │audit-chk │ │tax-plan │ │risk │ │corp-cons │ │oil-gas-val │ │ai-anomaly │
│budget-fc │ │forensic │ │revenue- │ │treasury││lbo-model │ │insurance-res│ │nlp-sent │
│statement │ │auto-recon│ │ recog │ │ │ │startup- │ │banking-comp │ │predictive │
│ -prep │ │ │ │ │ │ │ │ valuation│ │crypto-acct │ │ │
│invest- │ │ │ │ │ │ │ │debt-restr│ │variance-ana │ │ │
│ analysis │ │ │ │ │ │ │ │wacc-comp │ │product-prof │ │ │
│ │ │ │ │ │ │ │ │ecl-comp │ │earned-value │ │ │
│ │ │ │ │ │ │ │ │aro-comp │ │ │ │ │
│ │ │ │ │ │ │ │ │esg-report│ │ │ │ │
└────┬─────┘ └────┬─────┘ └────┬─────┘ └──┬───┘└────┬─────┘ └──────┬──────┘ └─────┬─────┘
│ │ │ │ │ │ │
└────────────┴─────┬──────┴───────────┴─────────┴──────────────┴──────────────┘
│
Skills cross-reference:
financial-analysis ↔ budget-forecast ↔ risk-assessment
audit-checklist ↔ forensic-accounting ↔ ai-anomaly-detection
statement-preparation ↔ corporate-consolidation ↔ revenue-recognition
oil-gas-valuation ↔ aro-computation ↔ ecl-computationSkill | Description |
Comprehensive financial statement analysis using ratio analysis, trend analysis, and performance metrics... | |
Build budgets, forecasts, and variance analysis models with rolling forecasts and scenario planning... | |
Prepare financial statements (Income Statement, Balance Sheet, Cash Flow) following IFRS/GAAP standards... | |
Evaluate investment opportunities using NPV, IRR, payback period, and sensitivity analysis... | |
Analyze product-line profitability, contribution margins, and cost allocation for decision-making... |
Skill | Description |
Generate comprehensive audit checklists, working papers, and test procedures for internal/external audits... | |
Detect fraud, analyze suspicious transactions, and conduct forensic investigations with data analytics... | |
Automate bank reconciliations, intercompany eliminations, and account reconciliations at scale... | |
Use machine learning to detect unusual transactions, outliers, and potential fraud in financial data... |
Skill | Description |
Optimize tax positions, plan tax strategies, and ensure compliance with local tax regulations... | |
Apply IFRS 15 / ASC 606 revenue recognition standards for contracts and performance obligations... | |
Prepare ESG (Environmental, Social, Governance) reports and sustainability disclosures... | |
Navigate Basel III, capital adequacy requirements, and banking regulatory compliance frameworks... |
Skill | Description |
Identify, quantify, and manage financial risks including credit, market, operational, and liquidity risks... | |
Optimize cash management, working capital, liquidity forecasting, and funding strategies... | |
Calculate Expected Credit Loss (ECL) under IFRS 9 for financial instruments and loan portfolios... |
Skill | Description |
Prepare consolidated financial statements for group entities with elimination entries... | |
Build leveraged buyout (LBO) models with sources & uses, debt schedules, and returns analysis... | |
Value early-stage companies using DCF, comparables, venture capital method, and scorecard approaches... | |
Analyze debt restructuring scenarios, workout agreements, and distressed company turnarounds... | |
Calculate Weighted Average Cost of Capital (WACC) for valuation and capital budgeting decisions... | |
Analyze venture debt facilities, warrants, and alternative financing structures for startups... | |
Compute Asset Retirement Obligations (ARO) for decommissioning and environmental liabilities... |
Skill | Description |
Value oil & gas reserves using NPV of future cash flows, reserve-based lending, and production forecasts... | |
Calculate insurance loss reserves, IBNR (Incurred But Not Reported), and actuarial estimates... | |
Account for cryptocurrency transactions, digital assets, NFTs, and DeFi protocols under emerging standards... | |
Analyze manufacturing cost variances (material, labor, overhead) for production cost control... | |
Track project financial performance using Earned Value Management (EVM) metrics and variance analysis... |
Skill | Description |
Extract sentiment and key insights from earnings calls, annual reports, and financial disclosures using NLP... | |
Forecast cash burn rates, runway, and funding requirements for startups and high-growth companies... |
Use npx skills to install skills directly:
# Install all skills
npx skills add GAJETOso/financeskills
# Install specific skills
npx skills add GAJETOso/financeskills --skill financial-analysis audit-checklist
# List available skills
npx skills add GAJETOso/financeskills --listThis automatically installs to your .agents/skills/ directory (and symlinks into .claude/skills/ for Claude Code compatibility).
Install via Claude Code's built-in plugin system:
# Add the marketplace
/plugin marketplace add GAJETOso/financeskills
# Install all finance skills
/plugin install finance-skillsClone the entire repo and copy the skills folder:
git clone https://github.com/GAJETOso/financeskills.git
cp -r financeskills/skills/* .agents/skills/Add as a submodule for easy updates:
git submodule add https://github.com/GAJETOso/financeskills.git .agents/financeskillsThen reference skills from .agents/financeskills/skills/.
Fork this repository
Customize skills for your specific industry/regulatory needs
Clone your fork into your projects
Use SkillKit to install skills across multiple AI agents (Claude Code, Cursor, Copilot, etc.):
# Install all skills
npx skillkit install GAJETOso/financeskills
# Install specific skills
npx skillkit install GAJETOso/financeskills --skill financial-analysis audit-checklist
# List available skills
npx skillkit install GAJETOso/financeskills --listOnce installed, just ask your agent to help with financial tasks:
"Analyze these financial statements and calculate key ratios"
→ Uses financial-analysis skill
"Prepare IFRS 15 revenue recognition journal entries for this contract"
→ Uses revenue-recognition skill
"Build a 5-year DCF model for this company"
→ Uses investment-analysis skill
"Calculate Expected Credit Loss for this loan portfolio"
→ Uses ecl-computation skill
"Prepare consolidated financial statements for our group"
→ Uses corporate-consolidation skillYou can also invoke skills directly:
/financial-analysis
/audit-checklist
/tax-planningfinancial-analysis - Ratio analysis, trend analysis, performance metrics
budget-forecast - Budgeting, forecasting, variance analysis
statement-preparation - Income Statement, Balance Sheet, Cash Flow
investment-analysis - NPV, IRR, investment evaluation
product-profitability - Product-line profitability analysis
audit-checklist - Audit procedures and working papers
forensic-accounting - Fraud detection and investigation
automated-reconciliation - Account reconciliation automation
ai-anomaly-detection - ML-based anomaly detection
tax-planning - Tax optimization and compliance
revenue-recognition - IFRS 15 / ASC 606 compliance
esg-reporting - Sustainability and ESG disclosures
banking-compliance - Basel III and banking regulations
risk-assessment - Financial risk identification and management
treasury-management - Cash and liquidity management
ecl-computation - IFRS 9 Expected Credit Loss
corporate-consolidation - Group consolidation
lbo-modeling - Leveraged buyout models
startup-valuation - Early-stage company valuation
debt-restructuring - Workout and restructuring analysis
wacc-computation - Cost of capital calculations
venture-debt - Alternative financing structures
aro-computation - Asset retirement obligations
oil-gas-valuation - Reserve valuation for energy sector
insurance-reserving - Insurance loss reserves and IBNR
crypto-accounting - Digital asset accounting
variance-analysis - Manufacturing cost variance analysis
earned-value-mgmt - Project financial performance tracking
nlp-earnings-sentiment - NLP for financial documents
predictive-burn-rate - Cash runway forecasting
This repository includes dedicated compliance mapping for:
IFRS Standards - International Financial Reporting Standards
US GAAP - Generally Accepted Accounting Principles
African Frameworks - Local regulatory requirements (Nigeria, South Africa, Kenya, Ghana)
Industry Regulations - Banking (Basel III), Insurance (IFRS 17), Oil & Gas (IFRS 6)
See Regulatory Compliance Mapping for detailed standard-to-skill mappings.
Get comfortable with the technical basics:
Python: Master Pandas, NumPy, and OpenPyXL for data manipulation
APIs: Connect to financial data sources (Yahoo Finance, Bloomberg, Alpha Vantage)
Git: Version control for financial models and collaboration
📚 Resources in /foundations
Learn to work effectively with AI agents:
Prompt Engineering: Craft precise instructions for complex financial reasoning
AI Tools: Leverage ChatGPT, Claude, GitHub Copilot, and Cursor for development
📚 Resources in /ai-usage
Apply specialized financial skills:
Start with foundational skills (financial-analysis, budget-forecast)
Progress to compliance skills (audit-checklist, revenue-recognition)
Master advanced topics (corporate-consolidation, predictive analytics)
📚 All skills available in /skills
Build end-to-end financial systems:
AI-powered Finance Assistant: Natural language interface for financial queries
Automated Reporting Tool: Dynamic generation of financial statements
Smart Audit System: AI-driven anomaly detection in transaction logs
📚 Project templates in /projects
financeskills/
├── foundations/ # Phase 1: Python, APIs, Git basics
├── ai-usage/ # Phase 2: Prompt engineering guides
├── skills/ # Phase 3: Modular skill definitions
│ ├── financial-analysis/
│ │ ├── SKILL.md # Skill instructions
│ │ ├── evals/ # Test cases
│ │ └── references/ # Accounting standards
│ ├── audit-checklist/
│ ├── tax-planning/
│ └── [40+ other skills]/
├── compliance/ # Regulatory mapping (IFRS, GAAP, local)
├── projects/ # Phase 4: Full-scale applications
└── tools/ # Helper utilities and integrationsSkills (auto-trigger + explicit): SKILL.md follows the open Agent Skills
spec - works in Claude Code, Codex CLI, Gemini CLI, Copilot/VS Code, Cursor,
Windsurf, and any tool reading .agents/skills/. In Claude Code, invoke
explicitly with /skill-name.
Slash commands: commands/ ships 52 commands - one per skill (full
coverage of all 43) plus workflow commands (/reconcile, /wacc,
/close-the-books, ...) with argument handling.
MCP connector: stdio (tools/mcp/server.py) or hosted HTTP
(tools/mcp/http_server.py + Dockerfile, see tools/mcp/DEPLOY.md) exposing 155+ tools (every calculator
across 38 skills, plus skill/standards/template readers) to ANY MCP client -
zero dependencies, no network, no keys. Config snippets in tools/mcp/README.md.
Install as a CLI: pipx install git+https://github.com/GAJETOso/financeskills
gives financeskills-mcp, financeskills-mcp-http, financeskills-validate,
financeskills-evals.
Live data: tools/connectors/ pulls Stripe and Plaid transactions into
skill-ready CSVs (each has a --mock mode). See tools/connectors/README.md.
Eval scorecard: see tools/evals/BASELINE.md. All 43 skills have file-based
evals with realistic fixtures (CSV/XLSX) and code-verified expected values.
Provenance: MANIFEST.sha256 checksums all content; skill scripts are
stdlib-only with no network access (see SECURITY.md).
Every skill is validated and testable:
Validator - python3 validate_skills.py enforces the full Agent Skills spec (frontmatter, evals schema, dead links, line limits), runs every script's self-test, and checks marketplace manifests against the skills on disk. CI runs it on every push.
Eval harness - python3 tools/evals/run_evals.py --all feeds each eval prompt (with its fixture files) to a model with the skill as system context and grades the assertions with an LLM judge. --dry-run validates schemas offline.
Deterministic scripts - 38 of 43 skills ship a scripts/calculate.py with self-tested functions, so agents compute with code instead of mental math.
Fixtures - flagship skills include realistic CSV fixtures (trial balances, lease portfolios, bank statements) with code-verified expected answers.
Templates - flagship skills ship output templates in assets/ for consistent memos and workpapers.
Found a way to improve a skill? Have a new skill to suggest? PRs and issues welcome!
See CONTRIBUTING.md for guidelines on adding or improving skills.
Add more industry-specific skills (real estate, healthcare, manufacturing)
Expand compliance coverage (SOX, GDPR, local African regulations)
Create more advanced analytics skills (time series forecasting, Monte Carlo simulation)
Build integration skills (QuickBooks, Xero, SAP, Oracle)
If you are an AI agent (Claude, Cursor, etc.), please refer to AGENTS.md for interaction guidelines and CLAUDE.md for project rules.
MIT - Use these however you want.
Built by Ayotunde Oso (aka GAJETOso) — Accountant in the oil & gas industry (Nigeria), with experience across IFRS, corporate finance, audit, and financial systems. Currently building AI-powered finance tools for accounting professionals in emerging markets.
Connect: GitHub
43 core financial skills
IFRS/GAAP compliance mapping
Industry-specific skills (oil & gas, banking, insurance)
AI agent integration (Claude Code, Cursor)
Interactive skill tutorials with Python notebooks
Integration with accounting software APIs (QuickBooks, Xero)
Video walkthroughs for each skill
Mobile app for on-the-go financial analysis
Community skill marketplace
Certification program for AI-powered accounting
Multi-language support (French, Portuguese for African markets)
Real-time collaboration features
Having issues? Questions? Want to contribute?
📧 Issues: Open an issue
💬 Discussions: Start a discussion
🤝 Contributing: See CONTRIBUTING.md
If you are an AI agent (Claude, Cursor, etc.), please refer to AGENTS.md for interaction guidelines and CLAUDE.md for project rules.
Built for accountants, by an accountant. Let's revolutionize financial analysis with AI.
[ai-anomaly-detection] iqr_anomalies
Inputs: values (required), k.
[ai-anomaly-detection] Flag points deviating from their trailing-window mean by > threshold sigmas.
Inputs: values (required), window, threshold.
[ai-anomaly-detection] Indices of values beyond threshold standard deviations from the mean.
Inputs: values (required), threshold.
[aro-computation] Year-by-year unwinding; closing balance equals the settlement amount.
Inputs: years (required), discount (required), opening_pv (required).
[aro-computation] PV of an inflated future restoration cost.
Inputs: years (required), discount (required), inflation (required), current_cost (required).
[aro-computation] IFRIC 1 revision: returns the adjustment to BOTH provision and asset (+ = increase).
Inputs: discount (required), years_remaining (required), old_remaining_pv (required), new_cost_estimate (required), +1 more.
[automated-reconciliation] Pair items with equal amount and reference. Each item used once.
Inputs: side_a (required), side_b (required).
[automated-reconciliation] Find a combination of candidates summing to the target amount (batched settlements).
Inputs: target (required), candidates (required), tol, max_combo.
[automated-reconciliation] Pair items whose amounts differ by <= tol; returns (a, b, difference).
Inputs: tol (required), side_a (required), side_b (required).
[banking-compliance] buffer_check
Inputs: cet1_ratio (required), dsib, minimum, conservation, +1 more.
[banking-compliance] Works for CET1, Tier 1, or Total capital against risk-weighted assets.
Inputs: rwa (required), capital (required).
[banking-compliance] lcr
Inputs: hqla (required), net_outflows_30d (required).
[banking-compliance] leverage_ratio
Inputs: tier1 (required), total_exposure (required).
[banking-compliance] nsfr
Inputs: required_stable_funding (required), available_stable_funding (required).
[banking-compliance] exposures: list of (exposure_amount, risk_weight).
Inputs: exposures (required).
[budget-forecast] cagr
Inputs: years (required), ending (required), beginning (required).
[budget-forecast] driver_based_budget
Inputs: price (required), units (required), fixed_costs (required), variable_cost_pct (required).
[budget-forecast] project_revenue
Inputs: base (required), growth_rates (required).
[budget-forecast] scenario_delta
Inputs: scenario (required), base_case (required).
[company-valuation] enterprise_value
Inputs: fcfs (required), wacc (required), terminal_value (required).
[company-valuation] equity_value
Inputs: ev (required), net_debt (required), minority_interest.
[company-valuation] NPV of cashflows starting at t=1.
Inputs: rate (required), cashflows (required).
[company-valuation] terminal_value_exit_multiple
Inputs: multiple (required), final_ebitda (required).
[company-valuation] terminal_value_gordon
Inputs: g (required), wacc (required), final_fcf (required).
[corporate-consolidation] Group share of equity - quick reasonableness check.
Inputs: pct_owned (required), parent_equity (required), sub_post_acq_profit (required), goodwill_impairment.
[corporate-consolidation] Full goodwill (IFRS 3 option / US GAAP requirement).
Inputs: consideration (required), nci_fair_value (required), net_identifiable_assets (required).
[corporate-consolidation] Partial goodwill: NCI measured at proportionate share of net assets.
Inputs: pct_acquired (required), consideration (required), net_identifiable_assets (required).
[corporate-consolidation] nci_at_acquisition
Inputs: method (required), pct_nci (required), nci_fair_value (required), net_identifiable_assets (required).
[credit-analysis] Headroom as fraction of covenant level. Positive = compliant.
Inputs: actual (required), covenant (required), lower_is_better.
[credit-analysis] Cash flow available for debt service / total debt service.
Inputs: cfads (required), debt_service (required).
[credit-analysis] (EBITDA - maintenance capex) / (interest + scheduled principal + leases).
Inputs: capex (required), ebitda (required), fixed_charges (required).
[credit-analysis] interest_coverage
Inputs: ebit (required), interest (required).
[credit-analysis] net_leverage
Inputs: cash (required), ebitda (required), total_debt (required).
[crypto-accounting] IAS 38 cost model: impair to lowest observed; no reversal above carrying (simplified - IFRS allows reversal to the impaired-cost ceiling).
Inputs: cost (required), current_price (required), lowest_observed (required).
[crypto-accounting] US GAAP (ASU 2023-08): measure at fair value, changes through net income.
Inputs: carrying (required), fair_value (required).
[crypto-accounting] lots: (units, unit_cost) in acquisition order.
Inputs: lots (required), sale_price (required), units_sold (required).
[cvp-breakeven] breakeven_sales
Inputs: cm_ratio (required), fixed_costs (required).
[cvp-breakeven] breakeven_units
Inputs: price (required), fixed_costs (required), variable_cost (required).
[cvp-breakeven] contribution_margin
Inputs: price (required), variable_cost (required).
[cvp-breakeven] degree_of_operating_leverage
Inputs: total_cm (required), operating_income (required).
[cvp-breakeven] Returned as a ratio of actual sales.
Inputs: actual_sales (required), breakeven_sales_value (required).
[cvp-breakeven] units_for_target_profit
Inputs: price (required), target (required), fixed_costs (required), variable_cost (required).
[debt-payoff] avalanche_schedule
Inputs: debts (required), monthly_budget (required).
[debt-payoff] interest_savings
Inputs: debts (required), monthly_budget (required).
[debt-payoff] monthly_interest
Inputs: balance (required), annual_rate (required).
[debt-payoff] snowball_schedule
Inputs: debts (required), monthly_budget (required).
[debt-restructuring] haircut_pct
Inputs: new_value (required), face_value (required).
[debt-restructuring] PV of interest-only payments plus bullet principal, at the discount rate.
Inputs: rate (required), years (required), principal (required), discount_rate (required).
[debt-restructuring] tranches: (name, claim) in priority order. Returns recovery per tranche.
Inputs: tranches (required), enterprise_value (required).
[deferred-tax] DTL for taxable differences, DTA for deductible ones, at enacted rate.
Inputs: tax_rate (required), temp_diff (required).
[deferred-tax] effective_tax_rate
Inputs: current_tax (required), pretax_income (required), deferred_tax_expense (required).
[deferred-tax] P&L (or OCI) impact of remeasuring the deferred balance at a new enacted rate.
Inputs: new_rate (required), old_rate (required), temp_diff (required).
[deferred-tax] Positive = taxable temporary difference (DTL); negative = deductible (DTA).
Inputs: tax_base (required), carrying_amount (required).
[earned-value-mgmt] EAC assuming remaining work at budgeted rates (variance was one-off).
Inputs: ac (required), ev (required), bac (required).
[earned-value-mgmt] EAC where both cost and schedule performance affect remaining work.
Inputs: ac (required), ev (required), bac (required), cpi (required), +1 more.
[earned-value-mgmt] EAC assuming current cost performance continues.
Inputs: bac (required), cpi (required).
[earned-value-mgmt] Cost performance needed on remaining work to hit BAC.
Inputs: ac (required), ev (required), bac (required).
[earned-value-mgmt] variances
Inputs: ac (required), ev (required), pv (required).
[ecl-computation] ECL for one period: PD x LGD x EAD x discount.
Inputs: ead (required), lgd (required), pd_ (required), discount_factor.
[ecl-computation] Lifetime ECL: sum of marginal-PD-weighted discounted losses per period (t starts at 1).
Inputs: eir (required), eads (required), lgds (required), marginal_pds (required).
[ecl-computation] Simplified-approach provision per aging bucket; includes 'total'.
Inputs: loss_rates (required), aging_balances (required).
[ecl-computation] Probability-weighted ECL across macro scenarios; weights must sum to 1.
Inputs: weights (required), scenario_ecls (required).
[financial-analysis] ROE = margin x asset turnover x leverage.
Inputs: assets (required), equity (required), revenue (required), net_income (required).
[financial-analysis] liquidity_ratios
Inputs: inventory (required), current_assets (required), current_liabilities (required).
[financial-analysis] profitability_ratios
Inputs: cogs (required), assets (required), equity (required), revenue (required), +1 more.
[financial-analysis] None when the denominator is zero/negative-meaningless rather than raising.
Inputs: den (required), num (required).
[financial-analysis] solvency_ratios
Inputs: ebit (required), equity (required), interest (required), total_liabilities (required).
[fixed-asset-accounting] Positive = gain, negative = loss.
Inputs: cost (required), proceeds (required), accumulated_depreciation (required).
[fixed-asset-accounting] Annual DDB charges; never depreciates below residual.
Inputs: cost (required), residual (required), life_years (required).
[fixed-asset-accounting] straight_line
Inputs: cost (required), residual (required), life_years (required).
[fixed-asset-accounting] units_of_production
Inputs: cost (required), residual (required), total_units (required), units_this_period (required).
[forensic-accounting] benford_test
Inputs: values (required).
[forensic-accounting] first_digit
Inputs: value (required).
[forensic-accounting] Share of amounts that are exact multiples of base - a fraud tell.
Inputs: values (required), base.
Return the full SKILL.md instructions for a skill (use list_skills for names).
Inputs: name (required).
Return a standards summary, e.g. family='ifrs', name='ifrs-16'. Summaries carry an as_of date - verify before decision-grade use.
Inputs: name (required), family (required).
Return an output template from a skill's assets/, e.g. skill='wacc-computation'.
Inputs: skill (required).
[insurance-reserving] Volume-weighted age-to-age factors from a cumulative paid/incurred triangle. triangle[i] = accident year i's cumulative values by development period.
Inputs: triangle (required).
[insurance-reserving] ibnr
Inputs: ultimates (required), paid_to_date (required).
[insurance-reserving] project_ultimates
Inputs: factors (required), triangle (required).
[intercompany-accounting] Eliminate IC revenue/COGS and defer unrealized margin in inventory.
Inputs: ic_revenue (required), unrealized_profit (required).
[intercompany-accounting] Compare IC receivable on A vs IC payable on B per counterparty key.
Inputs: entity_a_books (required), entity_b_books (required), tolerance.
[intercompany-accounting] Profit to eliminate on inventory still held within the group.
Inputs: cost (required), ic_sale_price (required), pct_still_held (required).
[inventory-costing] purchases: list of (units, unit_cost) in order. Returns COGS and ending value.
Inputs: purchases (required), units_sold (required).
[inventory-costing] IAS 2 lower of cost and NRV writedown (0 if NRV >= cost).
Inputs: nrv (required), cost (required).
[inventory-costing] lifo_cogs_periodic
Inputs: purchases (required), units_sold (required).
[inventory-costing] weighted_average_cogs
Inputs: purchases (required), units_sold (required).
[investment-analysis] Bisection IRR; requires a sign change over [lo, hi].
Inputs: cashflows (required), hi, lo, tol.
[investment-analysis] cashflows[0] at t=0 (usually negative).
Inputs: rate (required), cashflows (required).
[investment-analysis] Years to recover the t=0 outflow (fractional, undiscounted).
Inputs: cashflows (required).
[investment-analysis] profitability_index
Inputs: rate (required), cashflows (required).
[journal-entry] prorate_accrual
Inputs: days_elapsed (required), days_in_month (required), monthly_amount (required).
[journal-entry] Flip Dr/Cr for an accrual reversal.
Inputs: lines (required).
[journal-entry] lines: [{'account': str, 'dr': float, 'cr': float}]. Checks Dr = Cr and no dual-sided lines.
Inputs: lines (required).
[lbo-modeling] run_lbo
Inputs: a (required).
[lease-accounting] Year-by-year liability rollforward for payments in arrears.
Inputs: rate (required), payment (required), term_years (required).
[lease-accounting] PV of level lease payments at the discount rate (IBR or implicit rate).
Inputs: rate (required), payment (required), term_years (required), in_arrears.
[lease-accounting] Straight-line ROU depreciation (no residual).
Inputs: rou_asset (required), term_years (required).
List all FinanceSkills with their trigger descriptions.
Inputs: no named parameters.
List available accounting/auditing standards summaries (IFRS, IAS, US GAAP ASC, IPSAS, ISA, Basel, ESG...).
Inputs: no named parameters.
[net-worth-tracker] asset_allocation
Inputs: assets (required).
[net-worth-tracker] debt_to_asset_ratio
Inputs: assets (required), liabilities (required).
[net-worth-tracker] liquid_asset_ratio
Inputs: total_assets (required), liquid_assets (required).
[net-worth-tracker] net_worth
Inputs: assets (required), liabilities (required).
[net-worth-tracker] period_change
Inputs: current_nw (required), previous_nw (required).
[oil-gas-valuation] Annual production volumes under constant exponential decline.
Inputs: years (required), decline (required), initial_rate (required).
[oil-gas-valuation] netback
Inputs: price (required), royalty_pct (required), opex_per_unit (required).
[oil-gas-valuation] SEC-style PV-10: discounted net cashflows from year 1.
Inputs: volumes (required), netback_per_unit (required), rate.
[oil-gas-valuation] R/P life in years.
Inputs: reserves (required), annual_production (required).
[personal-budgeting] categorise_vs_target
Inputs: income (required), actual_needs (required), actual_wants (required), actual_savings (required).
[personal-budgeting] fifty_thirty_twenty
Inputs: monthly_income (required).
[personal-budgeting] savings_rate
Inputs: monthly_income (required), monthly_savings (required).
[personal-budgeting] zero_based_budget
Inputs: expenses (required), monthly_income (required).
[predictive-burn-rate] Convert invoicing into expected cash receipts via a lag distribution. curve[k] = share collected k months after invoicing; should sum to <= 1 (residual = bad debt).
Inputs: curve (required), invoiced_by_month (required).
[predictive-burn-rate] Months until cash drops below floor when burn grows at a monthly rate.
Inputs: cash (required), current_burn (required), monthly_growth (required), floor, +1 more.
[predictive-burn-rate] net_burn
Inputs: collections (required), gross_outflows (required).
[predictive-burn-rate] Months of runway at constant burn. Returns inf if cash-flow positive.
Inputs: cash (required), avg_net_burn (required).
[product-profitability] Allocate a shared cost pool pro-rata to a driver (revenue, units, hours...).
Inputs: drivers (required), shared_cost (required).
[product-profitability] product_pnl
Inputs: revenue (required), direct_costs (required), allocated_costs (required).
[product-profitability] unit_economics
Inputs: cac (required), arpu (required), variable_cost_per_user (required).
[retirement-planning] FV of lump sum + regular monthly contributions compounded monthly.
Inputs: years (required), principal (required), annual_rate (required), monthly_contribution.
[retirement-planning] nest_egg_target
Inputs: desired_annual_income (required), rate.
[retirement-planning] replacement_ratio
Inputs: retirement_income (required), pre_retirement_income (required).
[retirement-planning] Monthly contribution needed to reach target given existing principal.
Inputs: years (required), target (required), principal (required), annual_rate (required).
[retirement-planning] safe_withdrawal
Inputs: portfolio (required), rate.
[revenue-recognition] Allocate to performance obligations pro-rata to standalone selling prices.
Inputs: ssps (required), total_price (required).
[revenue-recognition] contract_liability
Inputs: billed (required), recognized (required).
[revenue-recognition] over_time_revenue
Inputs: allocated (required), total_months (required), months_elapsed (required).
[revenue-recognition] Split a deferred receipt into cash selling price (revenue) and interest.
Inputs: rate (required), years (required), deferred_price (required).
[risk-assessment] annualized_volatility
Inputs: periodic_returns (required), periods_per_year.
[risk-assessment] Loss threshold (positive number) at the given confidence, from history.
Inputs: returns (required), confidence.
[risk-assessment] minimum_variance_hedge_ratio
Inputs: corr (required), sigma_spot (required), sigma_future (required).
[risk-assessment] One-period Gaussian VaR in currency terms.
Inputs: mu (required), sigma (required), portfolio_value (required), confidence.
[startup-valuation] berkus
Inputs: team (required), rollout (required), prototype (required), sound_idea (required), +2 more.
[startup-valuation] Founder ownership after the round.
Inputs: new_money (required), post_money (required), founder_pct (required).
[startup-valuation] Post-money = exit / target return multiple; ownership = investment / post-money.
Inputs: exit_value (required), investment (required), target_multiple (required).
[startup-valuation] vc_method_irr
Inputs: years (required), exit_value (required), investment (required), required_irr (required).
[statement-preparation] cash_reconciliation
Inputs: cff (required), cfi (required), cfo (required), opening_cash (required).
[statement-preparation] financing_cash_flow
Inputs: debt_drawn (required), debt_repaid (required), dividends_paid (required), equity_issued.
[statement-preparation] investing_cash_flow
Inputs: capex (required), acquisitions, disposal_proceeds.
[statement-preparation] Deltas are (closing - opening). AR/inventory increases consume cash.
Inputs: net_income (required), depreciation (required), delta_payables (required), delta_inventory (required), +2 more.
[tax-efficient-investing] FV of a lump sum in a taxable account where gains are taxed annually.
Inputs: years (required), principal (required), annual_tax_rate (required), gross_annual_rate (required).
[tax-efficient-investing] Compare after-tax wealth from Roth vs. traditional contribution. Roth: contribute post-tax dollars, withdraw tax-free. Traditional: contribute pre-tax dollars, pay tax on withdrawal.
Inputs: years (required), contribution (required), future_tax_rate (required), current_tax_rate (required), +1 more.
[tax-efficient-investing] tax_drag_cost
Inputs: years (required), principal (required), annual_tax_rate (required), gross_annual_rate (required).
[tax-efficient-investing] Gross yield a taxable bond must offer to match a tax-free bond.
Inputs: tax_free_yield (required), marginal_tax_rate (required).
[tax-efficient-investing] FV of a lump sum in a tax-advantaged account (Roth / ISA) — no annual tax drag.
Inputs: years (required), principal (required), gross_annual_rate (required).
[tax-planning] PV saving from postponing a tax payment.
Inputs: tax_amount (required), discount_rate (required), years_deferred (required).
[tax-planning] effective_rate
Inputs: tax (required), income (required).
[tax-planning] marginal_rate
Inputs: income (required), brackets (required).
[tax-planning] brackets: (upper_bound, rate) ascending; last bound may be float('inf').
Inputs: income (required), brackets (required).
[three-statement-modeling] balance_check
Inputs: total_assets (required), total_equity (required), total_liabilities (required), tolerance.
[three-statement-modeling] cash_ties
Inputs: bs_cash (required), cf_closing_cash (required), tolerance.
[three-statement-modeling] Minimal linked projection. prior: revenue, cash, ppe, debt, equity, retained_earnings. assump: growth, ebitda_margin, da_pct_rev, capex_pct_rev, interest_rate, tax_rate, payout.
Inputs: prior (required), assump (required).
[three-statement-modeling] retained_earnings_rollforward
Inputs: dividends (required), net_income (required), opening_re (required).
[treasury-management] cash_position
Inputs: opening (required), receipts (required), disbursements (required).
[treasury-management] Covered interest parity: F = S x (1 + r_dom x t) / (1 + r_for x t).
Inputs: days (required), spot (required), foreign_rate (required), domestic_rate (required), +1 more.
[treasury-management] P&L of a forward purchase of foreign currency vs buying at maturity spot.
Inputs: notional_foreign (required), spot_at_maturity (required), contracted_forward (required).
[treasury-management] liquidity_runway_days
Inputs: daily_net_outflow (required), unrestricted_cash (required), committed_facilities (required).
[variance-analysis] fixed_overhead_variances
Inputs: foh_rate (required), actual_foh (required), budgeted_foh (required), std_hours_allowed (required).
[variance-analysis] label
Inputs: v (required).
[variance-analysis] std_hours = standard hours allowed for ACTUAL output.
Inputs: std_rate (required), act_hours (required), std_hours (required).
[variance-analysis] labor_rate_variance
Inputs: act_rate (required), std_rate (required), act_hours (required).
[variance-analysis] Positive = favorable.
Inputs: act_qty (required), act_price (required), std_price (required).
[variance-analysis] std_qty = standard quantity allowed for ACTUAL output. Positive = favorable.
Inputs: act_qty (required), std_qty (required), std_price (required).
[venture-debt] Monthly schedule: interest-only period then level amortization.
Inputs: io_months (required), principal (required), annual_rate (required), amort_months (required).
[venture-debt] runway_extension_months
Inputs: monthly_burn (required), net_loan_proceeds (required).
[venture-debt] total_cost_of_debt
Inputs: rows (required), closing_fee (required), final_payment_fee (required).
[wacc-computation] CAPM cost of equity. All inputs as decimals (0.05 = 5%).
Inputs: erp (required), beta (required), risk_free (required), size_premium, +1 more.
[wacc-computation] Hamada relevering at the target capital structure.
Inputs: tax_rate (required), debt_to_equity (required), unlevered_beta (required).
[wacc-computation] Hamada unlevering (debt beta assumed zero).
Inputs: tax_rate (required), levered_beta (required), debt_to_equity (required).
[wacc-computation] Weighted average cost of capital from market-value weights.
Inputs: i (required).
[working-capital-analysis] cash_conversion_cycle
Inputs: dio_v (required), dpo_v (required), dso_v (required).
[working-capital-analysis] dio
Inputs: cogs (required), avg_inventory (required), days.
[working-capital-analysis] dpo
Inputs: cogs (required), avg_payables (required), days.
[working-capital-analysis] dso
Inputs: revenue (required), avg_receivables (required), days.
[working-capital-analysis] Rough funding requirement implied by the cycle length.
Inputs: revenue (required), ccc_days (required), days.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 178 tools
Tools are organized by skill prefix, which helps distinguish across domains. Within a skill, individual tools are specialized and descriptions clarify intent. However, the large number of tools increases the chance of misselection, especially for similar-sounding tools like 'zscore_anomalies' and 'rolling_deviation_anomalies'.
Tools follow a consistent 'skill__tool_name' pattern using snake_case, which is predictable. Some tool names are vaguely named (e.g., 'label' in variance_analysis) or repeat the skill name (e.g., 'buffer_check'), but overall the pattern is maintained across all 178 tools.
178 tools is excessive for a single MCP server, even one covering multiple finance domains. This makes it difficult for an agent to efficiently navigate and select the right tool. The server would benefit from being split into smaller, domain-specific servers (e.g., 5-15 tools per domain).
The server covers an impressively broad range of finance and accounting topics, from anomaly detection to working capital analysis. Most common calculations (NPV, IRR, ratios, lease accounting, etc.) are present. A few areas are thin (e.g., crypto accounting has only two tools), but overall the surface is comprehensive for a financial analysis toolkit.
Open-source AI accounting skills verified by licensed accountants (tax, VAT, payroll).
AI agents for bookkeeping, reconciliation, and financial close for SMBs.
Primary-source SEC filing intelligence and financial/disclosure reconciliation for AI agents.
Connect AI agents to financial institution origination, analytics, and compliance workflows.
