Agent Blueprint

io.github.agent-blueprint由 GitHub 帳號 agent-blueprint 發布v0.7.23更新於 Nov 10, 2026

Client teams' blueprints for coding agents (token from an engagement). Free: agentblueprint.ai/mcp

發布方自己的服務:Agent Blueprint

概覽

AI 產生的概覽

將編碼代理接入 Agent Blueprint 登錄庫,用來建立業務概況、產生 AI 使用案例與藍圖,並追蹤實作進度。

可以做什麼
  • 讀取並更新組織的業務概況,包含技術概況與 AI 就緒度
  • 產生使用案例,啟動藍圖或完整流程作業並輪詢狀態
  • 列出並讀取藍圖摘要、商業論證、使用案例、實作計畫與規格中繼資料
  • 將藍圖下載為可供編碼代理使用的 Agent Skills 檔案目錄
  • 同步實作狀態並依預測目標回報實際指標
  • 重新計算商業論證財務資料並取得策略建議
適用情境
適合已與 Agent Blueprint 合作並取得 API 權杖的團隊,想讓編碼代理抓取藍圖、商業論證與計畫,或把實作進度回傳到登錄庫。
執行需求
透過 npx 以 stdio 在本機執行(需要 Node.js)。需要合作專案核發的 API 權杖,可透過 AGENT_BLUEPRINT_API_KEY 環境變數、權杖參數或已儲存的登入提供。需要連線至 Agent Blueprint API 的網路;API 位址可用 AGENT_BLUEPRINT_API_URL 覆寫。
安裝前請注意
API 權杖是與客戶合作綁定的憑證,應放在環境變數而非共用設定中。多個工具會寫入登錄庫:概況、使用案例、藍圖、商業論證與計畫的更新會建立版本快照並使下游產物過期,指標與實作狀態同步也會被記錄。業務概況與財務資料會傳送到廠商服務。

安裝

在 SourceWeft 中

  1. 開啟 儀表板中的 Agent Blueprint,將其新增到工作區。
  2. 為需要使用其工具的對話啟用該服務。

Desktop only,透過 STDIO。 STDIO 服務會啟動本機處理程序,因此需要 SourceWeft 桌面主機。

其他 MCP 客戶端

參照 儲存庫 中的啟動說明。

README

agentblueprint

CLI and MCP server for Agent Blueprint. Create structured business profiles, generate pipeline artifacts, explore blueprint data, sync implementation state, and download full Agent Skills directories. Vendor-agnostic output works with ServiceNow, Salesforce, OpenClaw, or any platform.

Agent Blueprint is delivered through hands-on engagements; API tokens are issued to client teams. If your organization works with Agent Blueprint, this package connects your coding agents to your blueprint registry.

Not a client? You don't need this package. Agent Blueprint also has free tools that need no account or token:

CLI Quick Start

bash
# Install globallynpm install -g agentblueprint
# Store your API token (one-time)agentblueprint login
# List blueprintsagentblueprint list
# Get a blueprint summary (JSON to stdout)agentblueprint get blueprint <id>
# Get other artifactsagentblueprint get business-case <id>agentblueprint get use-case <id>agentblueprint get implementation-plan <id>agentblueprint get implementation-spec <id>agentblueprint get business-profile
# Download as Agent Skills directoryagentblueprint download <id>
# Partner cross-org accessagentblueprint list --org <customer-org-id>agentblueprint get blueprint <id> --org <customer-org-id>

Or run without installing via npx:

bash
npx agentblueprint list --token <your-api-key>npx agentblueprint get blueprint <id> --token <your-api-key>

MCP Server Setup

The same binary auto-detects MCP mode when stdin is piped (non-interactive). No separate command needed.

Add to your Claude Code MCP config (.claude/settings.json or project settings):

json
{  "mcpServers": {    "agent-blueprint": {      "command": "npx",      "args": ["-y", "agentblueprint", "serve"],      "env": {        "AGENT_BLUEPRINT_API_KEY": "<your-api-key>"      }    }  }}

You can also start the MCP server explicitly with agentblueprint serve.

Get an API Key

API access comes with an Agent Blueprint engagement. If your organization is a client:

  1. Go to Agent Blueprint Settings > API Tokens
  2. Click "Create Token"
  3. Copy the token (shown once)

Not a client yet? Get in touch at agentblueprint.ai.

Supported Headless Workflow

Current scope is intentionally narrow:

existing PAT + existing org -> create_business_profile -> generate_use_cases -> generate_blueprint or trigger_full_pipeline -> get_generation_status -> download_blueprint

This release does not handle organization creation, OAuth bootstrap, or free-form context extraction into the business profile.

Download Blueprint as Agent Skills

Download a blueprint as a local Agent Skills directory that any coding agent can read from the filesystem. This is the recommended way to work with blueprints.

bash
# Using the CLI (after `agentblueprint login`)agentblueprint download <id>agentblueprint download <id> --dir ./my-skills
# Or via npxnpx agentblueprint download --token <key> --blueprint <id>

This creates an Agent Skills directory structure:

.agent-blueprint/<blueprint-slug>/├── SKILL.md                              # Overview + frontmatter (auto-discovered by agents)├── references/│   ├── business-context.md               # Use case, pain points, transformation story│   ├── agent-specifications.md           # Full agent specs with tools, guardrails, metrics│   ├── financial-case.md                 # ROI, cost breakdown, sensitivity, 5-year projection│   ├── implementation-roadmap.md         # Epics, stories, timeline, roles, dependencies│   ├── architecture-decisions.md         # Platform, pattern, integration gaps, feasibility│   └── guardrails-and-governance.md      # Risks, mitigation, per-agent guardrails└── scripts/    └── validate-spec.sh                  # Structure completeness checker

The Agent Skills standard is supported by Claude Code, Codex, Cursor, GitHub Copilot, Windsurf, and 18+ other coding agents. SKILL.md loads automatically at activation (~100 tokens), reference files load on demand.

Available Tools

ToolDescription
create_business_profileCreate or upsert a structured business profile for an existing organization
generate_use_casesGenerate normalized use cases from the current business profile and readiness assessment
generate_blueprintStart blueprint generation for a chosen use case and return an auditId
trigger_full_pipelineStart the full pipeline for an existing business profile and return a jobId
get_generation_statusPoll either blueprint generation or full-pipeline generation status
list_blueprintsList all blueprints (summaries)
get_blueprintBlueprint summary — title, agents, phases, pattern
get_business_caseBusiness case summary — ROI, pilot economics, recommendation
get_implementation_planImplementation plan summary — epics, timeline, story counts
get_use_caseUse case analysis for a blueprint
get_implementation_specImplementation spec metadata
get_business_profileOrganization business profile
download_blueprintDownload full blueprint as Agent Skills file manifest

The generate_blueprint and trigger_full_pipeline tools are asynchronous. Use get_generation_status to poll until the run completes, then call download_blueprint with the resolved blueprintId.

The get_blueprint, get_business_case, and get_implementation_plan tools return concise summaries optimized for agent context windows. For full details (agent specs, financial projections, user stories), use download_blueprint to get the complete Agent Skills directory.

Available Resources

URIDescription
agentblueprint://blueprintsBlueprint list (JSON)
agentblueprint://blueprints/{id}Blueprint detail (Markdown)
agentblueprint://blueprints/{id}/specImplementation spec (Markdown)

Authentication

Three ways to provide your API token (checked in this order):

  1. --token <key> flag on any command
  2. AGENT_BLUEPRINT_API_KEY environment variable
  3. agentblueprint login (saved to ~/.config/agentblueprint/config.json)

Environment Variables

VariableRequiredDefaultDescription
AGENT_BLUEPRINT_API_KEYNo—Your API token (alternative to agentblueprint login)
AGENT_BLUEPRINT_API_URLNohttps://app.agentblueprint.aiAPI base URL

來源:README.md,提交 8fc207a

工具

23

get_business_profile

Get the business profile for the organization. Returns company details, industry, strategic initiatives, technology profile, and AI readiness score.

輸入參數customerOrgId

create_business_profile

Create or upsert a structured business profile for an existing organization. Use this first in the headless workflow: existing PAT + existing org -> create_business_profile -> generate_use_cases or trigger_full_pipeline.

輸入參數fields必填customerOrgId

generate_use_cases

Generate AI use cases from the current business profile and readiness assessment for an existing organization. Returns normalized use case data with IDs you can pass to generate_blueprint.

輸入參數countguidanceguidanceTextcustomerOrgIdadditionalContextstrategicInitiativeId

generate_blueprint

Start blueprint generation for a chosen use case. This returns an auditId immediately; poll with get_generation_status until the blueprint is ready, then call download_blueprint.

輸入參數guidanceplatformuseCaseId必填guidanceTextassumeMissingcustomerOrgId

trigger_full_pipeline

Start the full pipeline for an existing business profile. This runs use cases through implementation plan generation asynchronously and returns a jobId; poll with get_generation_status.

輸入參數platformcustomerOrgIdbusinessProfileId必填specialInstructionsstrategicInitiativeId

get_generation_status

Poll either blueprint generation or full-pipeline generation status. Pass exactly one of auditId or jobId. When the run finishes with a blueprintId, call download_blueprint.

輸入參數jobIdauditIdcustomerOrgId

list_blueprints

List all blueprints for the organization. Returns summaries with id, title, platform, agent count, and lifecycle status.

輸入參數customerOrgId

get_blueprint

Get a blueprint summary by ID. Returns title, executive summary, agentic pattern, platform, agent names/roles, and phase overview. For full details, use download_blueprint.

輸入參數blueprintId必填customerOrgId

get_business_case

Get a business case summary for a blueprint. Returns executive summary, headline ROI numbers, pilot economics, and recommendation. For full financial analysis, use download_blueprint.

輸入參數blueprintId必填customerOrgId

get_implementation_plan

Get an implementation plan summary for a blueprint. Returns project overview, epic names with phases and story counts, and timeline. For full stories and dependencies, use download_blueprint.

輸入參數blueprintId必填customerOrgId

get_use_case

Get the use case analysis linked to a blueprint. Returns business challenge, success metrics, ROI estimate, and strategic alignment.

輸入參數blueprintId必填customerOrgId

get_implementation_spec

Get the compiled implementation spec for a blueprint. Returns metadata about the spec package including agent count, platform, and what artifacts are included.

輸入參數blueprintId必填customerOrgId

download_blueprint

Download a blueprint as an Agent Skills directory. Returns a JSON manifest with all files (SKILL.md, reference docs, scripts) that can be written to disk for any coding agent to consume. Use this instead of reading full blueprint/business-case/plan data via individual tools.

輸入參數platformblueprintId必填customerOrgId

sync_implementation_state

Sync implementation state back to Agent Blueprint. Reports which agents have been implemented, their status, platform artifacts, and any deviations from the spec. Each sync creates a versioned snapshot enabling progress tracking over time.

輸入參數stateData必填blueprintId必填customerOrgId

report_metric

Report actual performance metrics for a blueprint. The system auto-resolves predicted targets from the blueprint and returns deviation analysis. Supports multiple metrics in one call. Use this after implementing agents to track whether they hit their success criteria.

輸入參數metrics必填blueprintId必填customerOrgId

get_progress

Get implementation progress and performance metrics for a blueprint. Returns predicted targets, latest actual measurements with deviation analysis, and implementation state (if synced via sync_implementation_state). Use this to check how an implementation is tracking against the plan.

輸入參數blueprintId必填customerOrgId

update_blueprint

Update blueprint sections with actual implementation data. Sends full top-level sections that are shallow-merged into the existing blueprint. Creates a version snapshot before mutation and propagates staleness to business case and implementation plan. Use this when the implementation diverges from the original recommendation.

輸入參數sections必填blueprintId必填customerOrgId

update_business_case

Update business case sections. Shallow-merges provided sections into the existing business case data. Propagates staleness to implementation plan. After updating, consider calling recalculate_financials to refresh ROI projections.

輸入參數sections必填blueprintId必填customerOrgId

update_implementation_plan

Update implementation plan sections. Shallow-merges provided sections into the existing plan. This is a terminal artifact with no downstream staleness propagation.

輸入參數sections必填blueprintId必填customerOrgId

update_use_case

Update use case fields for a blueprint. Updates the use case linked to the specified blueprint. Propagates staleness to the blueprint.

輸入參數sections必填blueprintId必填customerOrgId

update_business_profile

Update business profile fields. Updates company details, technology profile, strategic initiatives, etc. Propagates staleness to use cases. Also syncs company name to the organization record.

輸入參數fields必填customerOrgId

recalculate_financials

Recalculate business case financials from current blueprint and profile data. Use this after updating the blueprint to refresh ROI projections, labor savings, and payback period. Clears staleness on the business case.

輸入參數blueprintId必填customerOrgId

get_recommendations

Get strategic recommendations for a blueprint based on implementation state, performance metrics, and AI landscape changes. Returns prioritized, actionable recommendations. Requires implementation state to be synced first via sync_implementation_state. Use ?force=true to regenerate.

輸入參數blueprintId必填customerOrgId

版本歷史

2
  1. v0.7.23最新Nov 3, 2026
  2. v0.7.15Sep 16, 2026

評分

已滿足適用檢查項的 90%

該評分衡量此收錄條目的完整程度,依據註冊中心條目、端點與公開來源計算得出。它不是對伺服器本身的評價,也不表示伺服器安全或目前可用:可連通性是帶日期的檢查結果,評分也不包含安裝量或使用者評價。

  • 近 30 天內可連通25 分
  • 已宣告認證方式10 分

    請在註冊中心條目中宣告憑證請求標頭,或保持端點能夠回應探測,以便判定其認證方式。

  • 已列出工具必需15 分
  • 描述必需15 分
  • 圖示10 分
  • 文件15 分
  • 授權條款10 分

其他 Agentblueprint 整合