Enables querying context from connected GitHub repositories, extracting code, documentation, and README files to provide relevant information for prompts.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@PromStack MCP Server find a prompt for writing a product description"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
PromStack의 공식 MCP(Model Context Protocol) 서버 구현체입니다.
**Official @modelcontextprotocol/sdk**를 사용하여 안정성과 호환성을 보장합니다.
이 서버를 통해 Claude Desktop, Cursor 등 MCP 호환 도구에서 PromStack의 프롬프트를 직접 불러오고 실행할 수 있습니다.
Node.js 18.0.0 이상
PromStack API Key (설정 > API Keys에서 발급)
npm install @promstack-1/mcp-server# 직접 실행
node bin/promstack-mcp.js --api-key YOUR_API_KEY
# 또는 환경 변수 사용
export PROMPTSTACK_API_KEY=YOUR_API_KEY
node bin/promstack-mcp.jsRelated MCP server: open-skills
Claude Desktop 설정 파일 로드:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"promstack": {
"command": "node",
"args": [
"D:/workspace/Prompt-Manager/mcp-server/bin/promstack-mcp.js",
"--api-key",
"YOUR_PK_LIVE_KEY"
]
}
}
}주의:
args의 경로는 실제mcp-server가 위치한 절대 경로로 수정해야 합니다.
list_prompts프로젝트의 프롬프트 목록을 조회합니다.
결과에는 각 프롬프트의 ID, 제목, 설명, 필요한 변수가 포함됩니다.
projectId를 지정하지 않으면 API Key에 연결된 모든 프로젝트의 프롬프트를 반환합니다.
Parameters:
projectId (optional): 특정 프로젝트의 프롬프트만 조회
search (optional): 검색어로 필터링
limit (optional): 최대 결과 수 (기본: 20, 최대: 100)
예시:
list_prompts({}) // 전체 프롬프트 목록
list_prompts({ projectId: 5 }) // 특정 프로젝트만
list_prompts({ search: "마케팅" }) // 검색get_prompt특정 프롬프트의 상세 정보를 조회합니다.
프롬프트 ID를 지정하면 해당 프롬프트의 전체 내용, 시스템 프롬프트, 필요한 변수 등을 반환합니다. 프롬프트를 실행하기 전에 상세 내용을 확인할 때 유용합니다.
Parameters:
promptId (required): 조회할 프롬프트 ID
예시:
get_prompt({ promptId: 42 })select_prompt작업 설명을 분석하여 가장 적합한 프롬프트를 자동으로 추천합니다.
키워드 매칭을 사용하여 작업 설명과 유사한 프롬프트를 찾습니다. 추천 결과에는 각 프롬프트의 관련성 점수와 사용법이 포함됩니다.
Selection Process:
작업 설명을 분석하여 키워드를 추출합니다
제목, 설명, 카테고리 매칭 순으로 가중치를 적용합니다
가장 관련성 높은 프롬프트를 반환합니다
중요: 첫 번째 추천 프롬프트(rank: 1)가 가장 적합한 선택입니다.
첫 번째 추천을 우선 사용하세요
여러 좋은 후보가 있으면 첫 번째를 선택하되, 다른 옵션도 참고 가능합니다
적합한 결과가 없으면 다른 키워드로 재검색하세요
Parameters:
taskDescription (required): 수행하려는 작업에 대한 설명
projectId (optional): 검색 범위를 특정 프로젝트로 제한
topK (optional): 추천할 프롬프트 수 (기본: 3, 최대: 10)
예시:
select_prompt({ taskDescription: "마케팅 이메일 작성" })
select_prompt({ taskDescription: "API 인증 구현", projectId: 5 })export_skill프롬프트를 Claude Skills 형식(SKILL.md)으로 내보냅니다.
출력에는 YAML frontmatter(name, description)와 Markdown 본문이 포함됩니다. 이 형식은 Claude Desktop, Claude Code 등에서 스킬로 사용할 수 있습니다.
⚠️ 유료 기능: Plus/Pro 플랜에서만 사용 가능합니다.
Parameters:
promptId (required): 내보낼 프롬프트 ID
includeResources (optional): 참조 리소스 포함 여부 (기본: false)
예시:
export_skill({ promptId: 42 })
export_skill({ promptId: 42, includeResources: true })query_context 🆕프롬프트에 연결된 GitHub 저장소 또는 문서 URL에서 컨텍스트를 검색합니다.
연결된 소스에서 코드, 문서, README 등을 추출하여 반환합니다. 결과는 캐싱되어 반복 요청 시 빠르게 응답합니다 (TTL: 1시간).
특징:
10,000자 제한: Context7 호환 형식
자동 캐싱: 1시간 TTL로 반복 요청 최적화
다중 소스: GitHub 저장소 및 문서 URL 동시 지원
Parameters:
promptId (required): 프롬프트 ID
maxTokens (optional): 최대 토큰 수 (기본: 10000, 최대: 10000)
refreshCache (optional): 캐시 무시하고 새로 가져오기 (기본: false)
Returns:
연결된 소스에서 추출한 컨텍스트 (최대 10,000자)
캐시 상태 정보 (cached: true/false)
소스별 문자 수 통계
예시:
query_context({ promptId: 42 })
query_context({ promptId: 42, maxTokens: 5000 })
query_context({ promptId: 42, refreshCache: true })이 서버는 Proxy 역할을 수행합니다.
User -> MCP Client (Claude) -> MCP Server (Local) -> HTTP (JSON-RPC) -> PromStack BackendSDK: @modelcontextprotocol/sdk
Transport: StdioServerTransport
Validation: zod
| Name | Required | Description | Default |
|---|---|---|---|
| promptId | Yes | ID of the prompt to export as a skill | |
| includeResources | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| promptId | Yes | ID of the prompt to retrieve |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| projectId | No | Filter by Project ID | |
| search | No | Search term for prompts | |
| limit | No | Limit number of results |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| promptId | Yes | ID of the prompt to query context for | |
| maxTokens | No | Maximum tokens (max: 10000) | |
| refreshCache | No | Force refresh cache |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
| Name | Required | Description | Default |
|---|---|---|---|
| taskDescription | Yes | Description of the task to find prompts for | |
| projectId | No | ||
| topK | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Tool has no description.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Tool has no description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Tool has no description.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Tool has no description.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Tool has no description.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Tool has no description.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
export_skillget_promptlist_promptsquery_contextselect_promptThe tools have distinct names but their purposes are unclear due to missing descriptions, making it hard to assess true overlap. 'list_prompts' and 'select_prompt' might overlap in functionality, while 'query_context' and 'get_prompt' could be ambiguous without context. The lack of descriptions forces reliance on naming alone, which introduces uncertainty.
All tool names follow a consistent verb_noun pattern with snake_case, such as 'export_skill', 'get_prompt', and 'list_prompts'. There are no deviations in naming conventions, making the set predictable and easy to parse for an agent.
With 5 tools, the count is reasonable for a server focused on prompts and skills, as inferred from the names. It's slightly lean but not overly sparse, suggesting a well-scoped set that could cover core operations without being overwhelming.
Inferred domain involves managing prompts and skills, but the tool set lacks clear CRUD coverage—there's no 'create_prompt', 'update_prompt', or 'delete_prompt', and 'export_skill' stands alone without complementary tools. The missing descriptions exacerbate gaps, making it likely agents will encounter dead ends in workflows.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Talk to your public-facing AI from any MCP client — Claude, ChatGPT, Cursor, Cline, Windsurf.
Professional prompt library over remote MCP: 13 verticals, free discovery scope.
Self-hosted AI prompt library: prompts, collections, tags, teams, chains. 29 MCP tools for agents.
Sharebench — search & pull AI skills, agents, prompts & playbooks (SKILL.md) into any MCP client
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/PromStack-1/mcp-server'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
