What does this skill do, and when should you use it?
This skill builds an accrual schedule from an entity, accounting period, and firm accrual-policy list. For each policy item, it calculates the basis, period portion, amount already booked, current-period accrual, and supporting reference. It drafts a journal entry for every non-zero accrual. The entries are staged for controller approval and are not posted to the ledger.
It reads the entity, period, accrual-policy list, and prior accruals plus current-period invoices posted for each item from the internal-gl MCP. It derives the accrual name, contractual or estimated basis, period allocation, current-period accrual, and support reference. Basis sources may include an engagement letter, compensation plan, or trailing-three-month average. For each non-zero amount, it drafts a debit to the expense account and a credit to accrued liabilities, with a memo; auto-reversing items receive a note that they reverse on day 1 of the next period.
- A fund-finance professional needs a structured accrual schedule during the month-end close.
- A controller wants to review the calculation basis, support references, and draft entries before sign-off.
- An accounting team needs to combine prior accruals with current-period invoices to determine the required adjustment.
- A finance-operations user needs consistent schedules for items such as audit fees, bonuses, and utilities.
How do you install this skill?
- Do not automatically publish or post the draft JE; a controller should verify policy, support, accounts, period, and amounts.
- Before deployment, validate the internal-gl MCP permission scope, data flow, availability, and mainland-China reachability, and add failure handling and sensitive-data controls.
- Add tests and clear exceptions for duplicate invoices, missing support, cross-period items, negatives, currencies, rounding, estimate changes, and non-zero differences.
- MCP Server
internal-gl MCP
In Cowork, open Settings → Plugins → Add plugin, paste https://github.com/anthropics/financial-services, and select Month-End Closer from the marketplace list. The README does not document installing this individual skill separately.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/anthropics/financial-services.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/plugins/agent-plugins/month-end-closer/skills/accrual-schedule" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Build the June 2026 month-end accrual schedule for Entity X using this policy list: audit fee, bonus, and utilities.
Provide an entity, accounting period, and firm accrual-policy list, then request an accrual schedule. Example: "Build the June 2026 month-end accrual schedule for Entity X using this policy list: audit fee, bonus, and utilities." The source does not document the exact trigger syntax or internal-gl MCP setup.
What are this skill's strengths and limitations?
- Produces a clear schedule covering calculations, support references, and draft entries.
- Separates the period portion from amounts already booked.
- Keeps journal entries staged for controller approval rather than posting them.
- Handles the memo requirement for auto-reversing accruals.
- Requires a firm accrual-policy list and internal-gl MCP data.
- Supporting invoices and vendor statements are explicitly treated as untrusted and require review.
- Does not post or approve journal entries.
- The source provides no test suite, error-handling details, or standalone installation procedure.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Period-End Accrual Planner this page ✓ Anthropic · Official | 48 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| Finance & Accounting Super-Skill | 46 · Use with care | ★ 26 | 7mo ago | — |
| Month-End Variance Commentary ✓ Anthropic · Official | 42 · Not recommended | ★ 40k | 1mo ago | Apache-2.0 |
| Reconciliation Break Tracer ✓ Anthropic · Official | 53 · Use with care | ★ 40k | 1mo ago | Apache-2.0 |
| Nine-Chapter Equity Research Skill | 63 · Recommended | ★ 462 | 2mo ago | MIT |
How did FollowSkills review this skill?
The skill treats invoices and vendor statements as untrusted inputs, requires support references, and explicitly limits the output to a controller-reviewed draft that must not be posted, reducing external-effect risk. Points are deducted because least privilege, sensitive financial-data handling, MCP data flows, explicit user confirmation, and rollback or audit controls are not specified.
The workflow and field definitions are internally coherent and include formulas, support references, and an auto-reversal note. Points are deducted because the internal-gl MCP dependency lacks availability, interface, and failure guidance, while missing-data, duplicate-booking, negative, cross-period, zero-value, and formula-exception cases are not addressed; static evidence cannot support a score above 10.
The audience, month-end-close scenario, entity/period/policy-list inputs, and schedule plus JE-draft outputs are reasonably clear, with direct posting excluded. Points are deducted because trigger exclusions, complex-policy boundaries, non-English support, and mainland-China MCP reachability are not documented; local configuration is still required.
The SKILL.md is concise and reasonably layered; the repository README, Apache-2.0 license, plugin-validation CI, and contribution process provide some governance evidence. Points are deducted because the skill itself lacks installation/dependency notes, examples, FAQs, versioning, changelog, named maintenance responsibility, and an explicit update path.
The skill can transform a policy list into itemized calculations, support references, and draft entries, with output staged for controller review, so the core task is defined. Points are deducted because no representative output or execution evidence is provided, and account mapping, policy conflicts, rounding, tax, currency, and cutoff judgments remain uncovered and may require substantial review; static calibration caps this at 7.
The formulas, source-field definitions, and support-reference requirement provide limited auditability, and repository CI can statically validate plugin structure. Points are deducted because there is no key-path test suite, execution record, or independent third-party corroboration demonstrating reproducible calculations; static review caps this at 5.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →