mcp
Research-backed linting + generation for agent context files (CLAUDE.md, AGENTS.md, Cursor rules).
使うべきか
品質と安全性
ツール定義とプロトコルへの準拠に関する自動分析に基づいています。
コンテキストコスト
これは、サーバーのツールがモデルのコンテキストに読み込まれるたびに消費されるおおよそのトークン数です。数が多いほど、ほかのタスクに使える注意が減ります。
インストール
ワンクリックインストール
これを `claude_desktop_config.json` ファイルに追加してください:
{
"mcpServers": {
"mcp": {
"url": "https://mcp.promptarch.ai/mcp"
}
}
}リモートエンドポイント
https://mcp.promptarch.ai/mcpstreamable-httpできること
ツール一覧
ツール（3）
🟢promptarch_lint_artifact(content, format, filename)
Lint an AI agent context file (CLAUDE.md, AGENTS.md, Cursor rules, Copilot instructions, memory files) with PromptArch's deterministic linter: ~30 research-backed rule families covering current-model anti-patterns (tool-forcing, reasoning-echo, sycophancy triggers, contradictions), per-tool size budgets, secrets/dangerous commands, and hidden-Unicode injection ("Rules File Backdoor"). Returns grade, 0-100 score, and line-anchored findings. Free: no account or credits needed.
入力スキーマ
{
"type": "object",
"properties": {
"content": {
"type": "string",
"description": "The full file content to lint."
},
"format": {
"description": "Override format auto-detection.",
"type": "string",
"enum": [
"claude_md",
"agents_md",
"cursor_mdc",
"copilot_instructions",
"memory_file"
]
},
"filename": {
"description": "Filename hint for format detection (e.g. CLAUDE.md).",
"type": "string"
}
},
"required": [
"content"
],
"$schema": "http://json-schema.org/draft-07/schema#"
}🟢promptarch_list_artifact_types
List the artifact types promptarch_generate_artifact can produce from a project description.
入力スキーマ
{
"type": "object",
"properties": {},
"$schema": "http://json-schema.org/draft-07/schema#"
}⚪promptarch_generate_artifact(artifact, project_name, project_description, tech_stack, repo_structure, ...)
Generate an AI agent context artifact (e.g. context_pack, claude_md) from a project description. Requires a PromptArch API key configured as an Authorization: Bearer pk_... header on this MCP server. Consumes credits.
入力スキーマ
{
"type": "object",
"properties": {
"artifact": {
"type": "string",
"enum": [
"context_pack",
"claude_md",
"agents_md",
"cursor_rule"
],
"description": "Artifact id. See promptarch_list_artifact_types. Other Studio artifact types need artifact-specific inputs this tool does not accept."
},
"project_name": {
"type": "string",
"maxLength": 1900
},
"project_description": {
"type": "string",
"maxLength": 1900
},
"tech_stack": {
"type": "string",
"maxLength": 1900
},
"repo_structure": {
"description": "Directory tree or structure summary. Max 1900 chars: summarize a large repo rather than pasting a full tree.",
"type": "string",
"maxLength": 1900
},
"commands": {
"type": "string",
"maxLength": 1900
},
"coding_conventions": {
"description": "Conventions, invariants, house rules. Max 1900 chars.",
"type": "string",
"maxLength": 1900
},
"target_model": {
"description": "Target model family (default Claude).",
"type": "string",
"maxLength": 100
}
},
"required": [
"artifact",
"project_name",
"project_description"
],
"$schema": "http://json-schema.org/draft-07/schema#"
}コミュニティ
エビデンス