Skip to main content
plugin

month-end-close — Claude Code plugin: what it does, how to install

Coordinates financial close calendars, balances reconciliations, and audits variance drivers. Runs Python stdlib scripts locally on CSV and JSON files.

Install into Claude Code · Plugin marketplace All methods and platforms ↓
/plugin marketplace add alirezarezvani/claude-skills
/plugin install month-end-close@claude-code-skills
Open on GitHub

What month-end-close does

The plugin bundles the cs-close-controller agent and three Python tools. It requires no network connection. The close_calendar_planner.py script arranges task dependencies and isolates the critical path. Execution halts with an error code if a key control lacks an independent reviewer or if tasks push the close past the target workday.

Account proofs run through reconciliation_matcher.py, which pairs general ledger records with sub-ledgers or bank data by reference and amount. It prepares adjusted balance schedules, flags aging items, and stops when differences remain unexplained. The third tool, flux_analyzer.py, audits balance swings against user materiality rules, rejecting explanations that omit quantified drivers.

All scripts run strictly offline on exported CSV or JSON files. Materiality thresholds and aging limits depend entirely on the parameters you pass to the commands.

What it helps with

  • calculate the critical path and task slack in a close calendar
  • reconcile general ledger balances against bank statements
  • detect self-reviewed controls and unassigned close tasks
  • block boilerplate variance explanations lacking numeric drivers

Example prompts for the agent

  • Run the complete close workflow: /cs-month-end-close close_plan.json gl_cash.csv bank.csv tb_compare.csv
  • Inspect account fluctuations with set thresholds: /cs-flux-review tb_compare.csv --pct 10 --abs 10000
  • Run sample data through the reconciliation script: python scripts/reconciliation_matcher.py --sample

How to install month-end-close

Plugin marketplace

/plugin marketplace add alirezarezvani/claude-skills
/plugin install month-end-close@claude-code-skills

Step by step, with commands for each platform: Claude Code plugins: marketplaces, installing, and building your own

Questions about month-end-close

How do I install month-end-close in Claude Code?

Add the marketplace source with /plugin marketplace add alirezarezvani/claude-skills. Next, install the package using /plugin install month-end-close@claude-code-skills.

What does month-end-close do?

It maps task dependencies for accounting close dates, matches ledger lines to bank proofs, and validates flux variance drivers.

Does month-end-close need access to an ERP or external APIs?

No, it uses standard Python libraries and runs locally. You only need to export your accounting reports to CSV or JSON.

Source text

month-end-close · MIT
# month-end-close

> A slow close is rarely slow everywhere. One chain of tasks sets the date,
> reconciliations "agree" without a proof, and "timing" passes as a reason.

Three deterministic scripts plan the close on its critical path, prove each
reconciliation, and refuse flux explanations that do not name a driver.
No ERP connection, no network: export CSV or JSON and run.

**Not accounting or audit advice.** Thresholds and aging limits are your policy.

## Install

```bash
/plugin marketplace add alirezarezvani/claude-skills
/plugin install month-end-close
```

## Use

```bash
/cs-month-end-close close_plan.json gl_cash.csv bank.csv tb_compare.csv   # full pass
/cs-flux-review tb_compare.csv --pct 10 --abs 10000                       # just flux
```

Or run the scripts directly. Each has `--help`, `--sample` and `--output json`:

```bash
cd skills/month-end-close
python scripts/close_calendar_planner.py --sample
python scripts/reconciliation_matcher.py --sample
python scripts/flux_analyzer.py --sample
```

## The three scripts

| Script | Question | Exit codes |
|---|---|---|
| `close_calendar_planner.py` | Which tasks set the close date? Topological order, critical path and slack, WD+n deadlines mapped to dates, owner load per day, **key controls without an independent reviewer** | 0 · 2 at risk · 3 bad input or cycle · **4 blocked** |
| `reconciliation_matcher.py` | Does it prove? Match by reference then amount and date, adjusted-balance proof, transposition hints, aging buckets, journal entries needed, **preparer ≠ reviewer** | 0 · 2 items · 3 · **4 unreconciled** |
| `flux_analyzer.py` | Is every material change explained? Your threshold rule, sign flips, new or closed balances, **explanation lint** (missing / boilerplate / no driver / no evidence) | 0 · 2 evidence · 3 · **4 incomplete** |

## What it cites

COSO 2013 Principles 10 and 12; PCAOB AS 2201, AS 2305 and Staff Audit Practice
Alert 11; a published university reconciliation policy for aging buckets; and
APQC days-to-close benchmarks, labelled by how directly they were verified.
Every claim in `skills/month-end-close/references/` carries a confidence marker.

## Distinct from

- `financial-analyst` — ratios, DCF, budgets, forecasts.
- `saas-metrics-coach` — ARR, churn and unit economics.
- `research-finance` — R&D program budgets and capitalisation.

## License

MIT. See the repository [LICENSE](../../LICENSE).

© Alireza Rezvani · MIT · alirezarezvani/claude-skills

Similar plugins

Everything for Finance and business →

From the same set463

The whole set →
red-green-proof Writes a test that fails on the current code, applies the fix, then reverts it to confirm the test goes red again. Built for debugging and incident work. 3installs swedish-mentor This skill mentors Swedish learners: it places you at a CEFR level and picks YouTube clips and podcast episodes for the skill you want to train. Useful when you need a plan, not random videos. 2installs coding-agent-rollout Audits Claude settings before a team rollout: lints permissions across scopes, inventories MCP servers and plugins, and turns local session data into adoption counts. 1installs ai-decision-authority Sorts decisions into three AI governance zones by stakes and reversibility, so you know where a human signs off and where the agent just acts. ★ 28,733repo stars coding-agent-rollout This plugin audits Claude Code settings, MCP and plugin inventories, and CLAUDE.md layers. It also tracks a team’s rollout stage from local session counts. ★ 28,733repo stars agent-regression-evals Compares agent runs before and after a change and decides whether the merge is safe. Useful when one trial can't separate a regression from noise. ★ 28,733repo stars loop-library Guides you through designing, auditing, and finding repeatable AI agent loops with explicit guardrails and stop conditions. Helpful for managing Claude Code autonomy. ★ 28,733repo stars spinning-up-deep-rl A knowledge base built from Spinning Up in Deep RL: a resident core with the algorithm taxonomy plus 20 chapters loaded on demand. ★ 28,733repo stars

Project sponsors

Companies that keep this analytics open
Gold sponsor

Impressions, clicks and code copies — open statistics. We keep the numbers public for the same reason NeBlask itself is public. Become a sponsor