shiwake-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| screen_journalsA | 仕訳データに全ルールを適用し、リスクスコアの高い順に並べ替えて返す。先に人間が見るべき束を絞り込むための一次スクリーニング。検出は不正の証拠ではない。 |
| check_balanceA | 貸借が一致しない仕訳だけを返す。取込不良と手入力の混入を最初に落とすために使う。 |
| benford_analysisA | 金額の先頭桁の分布をベンフォードの法則と比較し、MAD と χ² を返す。母集団の性質を見るための道具で、個別仕訳の判定には使えない。 |
| detect_duplicatesB | 計上日・金額・借方科目・貸方科目が完全に一致する仕訳をグループにして返す。 |
| list_rulesA | 実装されているスクリーニングルールの一覧と、それぞれが何を示すかの説明を返す。 |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 5 tools
Each tool targets a distinct screening concern—aggregate risk, balance errors, Benford distribution, duplicates, and rule metadata. screen_journals conceptually overlaps by applying all rules, but its description clearly frames it as a first-pass triage rather than a replacement for the focused checks.
Four tools follow a clear verb_noun pattern (screen_journals, check_balance, detect_duplicates, list_rules), and all names are snake_case. benford_analysis breaks the pattern by using a noun phrase instead of a verb-object form, though it remains readable.
Five tools is a well-scoped size for a specialized journal-entry screening server. Each tool covers a meaningful stage of the workflow without redundancy or bloat.
The set covers first-pass screening, unbalanced-entry cleanup, population-level Benford testing, duplicate detection, and rule documentation, which forms a coherent lifecycle. A per-entry rule explanation or a way to inspect raw risk factors would be a minor gap, but agents can work around it via screen_journals and list_rules.