跳到正文
cccskills
免费在 GitHub 公开

pipedrive-automation

Manage Pipedrive sales records through Rube MCP, a connector for agent tools. Create and update deals, contacts, and organizations, schedule activities, attach notes, and review pipelines and stages to track sales progress.

中文简介正在准备中。当前显示英语简介。

原文查看原文说明

Automate Pipedrive CRM operations including deals, contacts, organizations, activities, notes, and pipeline management via Rube MCP (Composio). Always search tools first for current schemas.

查看安装方法

适用场景

  • Create and update deals linked to contacts and companies
  • Search and maintain contact and organization records
  • Schedule calls and meetings linked to deals
  • Add notes and review deals by sales stage

包含的文件(1)

  • SKILL.md12.0 KB

SKILL.md(原文)

安装前可以查看交给 AI 代理的指令内容。

Pipedrive Automation via Rube MCP

Automate Pipedrive CRM workflows including deal management, contact and organization operations, activity scheduling, notes, and pipeline/stage queries through Composio's Pipedrive toolkit.

Toolkit docs: composio.dev/toolkits/pipedrive

Prerequisites

  • Rube MCP must be connected (RUBE_SEARCH_TOOLS available)
  • Active Pipedrive connection via RUBE_MANAGE_CONNECTIONS with toolkit pipedrive
  • Always call RUBE_SEARCH_TOOLS first to get current tool schemas

Setup

Get Rube MCP: Add https://rube.app/mcp as an MCP server in your client configuration. No API keys needed — just add the endpoint and it works.

  1. Verify Rube MCP is available by confirming RUBE_SEARCH_TOOLS responds
  2. Call RUBE_MANAGE_CONNECTIONS with toolkit pipedrive
  3. If connection is not ACTIVE, follow the returned auth link to complete Pipedrive OAuth
  4. Confirm connection status shows ACTIVE before running any workflows

Core Workflows

1. Create and Manage Deals

When to use: User wants to create a new deal, update an existing deal, or review deal details in the sales pipeline.

Tool sequence:

  1. PIPEDRIVE_SEARCH_ORGANIZATIONS - Find existing org to link to the deal [Optional]
  2. PIPEDRIVE_ADD_AN_ORGANIZATION - Create organization if none found [Optional]
  3. PIPEDRIVE_SEARCH_PERSONS - Find existing contact to link [Optional]
  4. PIPEDRIVE_ADD_A_PERSON - Create contact if none found [Optional]
  5. PIPEDRIVE_GET_ALL_PIPELINES - Resolve pipeline ID [Prerequisite]
  6. PIPEDRIVE_GET_ALL_STAGES - Resolve stage ID within the pipeline [Prerequisite]
  7. PIPEDRIVE_ADD_A_DEAL - Create the deal with title, value, org_id, person_id, stage_id [Required]
  8. PIPEDRIVE_UPDATE_A_DEAL - Modify deal properties after creation [Optional]
  9. PIPEDRIVE_ADD_A_PRODUCT_TO_A_DEAL - Attach line items/products [Optional]

Key parameters:

  • title: Deal title (required for creation)
  • value: Monetary value of the deal
  • currency: 3-letter ISO currency code (e.g., "USD")
  • pipeline_id / stage_id: Numeric IDs for pipeline placement
  • org_id / person_id: Link to organization and contact
  • status: "open", "won", or "lost"
  • expected_close_date: Format YYYY-MM-DD

Pitfalls:

  • title is the only required field for PIPEDRIVE_ADD_A_DEAL; all others are optional
  • Custom fields appear as long hash keys in responses; use dealFields endpoint to map them
  • PIPEDRIVE_UPDATE_A_DEAL requires the numeric id of the deal
  • Setting status to "lost" requires also providing lost_reason

2. Manage Contacts (Persons and Organizations)

When to use: User wants to create, update, search, or list contacts and companies in Pipedrive.

Tool sequence:

  1. PIPEDRIVE_SEARCH_PERSONS - Search for existing person by name, email, or phone [Prerequisite]
  2. PIPEDRIVE_ADD_A_PERSON - Create new contact if not found [Required]
  3. PIPEDRIVE_UPDATE_A_PERSON - Modify existing contact details [Optional]
  4. PIPEDRIVE_GET_DETAILS_OF_A_PERSON - Retrieve full contact record [Optional]
  5. PIPEDRIVE_SEARCH_ORGANIZATIONS - Search for existing organization [Prerequisite]
  6. PIPEDRIVE_ADD_AN_ORGANIZATION - Create new organization if not found [Required]
  7. PIPEDRIVE_UPDATE_AN_ORGANIZATION - Modify organization properties [Optional]
  8. PIPEDRIVE_GET_DETAILS_OF_AN_ORGANIZATION - Retrieve full org record [Optional]

Key parameters:

  • name: Required for both person and organization creation
  • email: Array of objects with value, label, primary fields for persons
  • phone: Array of objects with value, label, primary fields for persons
  • org_id: Link a person to an organization
  • visible_to: 1 = owner only, 3 = entire company
  • term: Search term for SEARCH_PERSONS / SEARCH_ORGANIZATIONS (minimum 2 characters)

Pitfalls:

  • PIPEDRIVE_ADD_AN_ORGANIZATION may auto-merge with an existing org; check response.additional_data.didMerge
  • Email and phone fields are arrays of objects, not plain strings: [{"value": "test@example.com", "label": "work", "primary": true}]
  • PIPEDRIVE_SEARCH_PERSONS wildcards like * or @ are NOT supported; use PIPEDRIVE_GET_ALL_PERSONS to list all
  • Deletion via PIPEDRIVE_DELETE_A_PERSON or PIPEDRIVE_DELETE_AN_ORGANIZATION is soft-delete with 30-day retention, then permanent

3. Schedule and Track Activities

When to use: User wants to create calls, meetings, tasks, or other activities linked to deals, contacts, or organizations.

Tool sequence:

  1. PIPEDRIVE_SEARCH_PERSONS or PIPEDRIVE_GET_DETAILS_OF_A_DEAL - Resolve linked entity IDs [Prerequisite]
  2. PIPEDRIVE_ADD_AN_ACTIVITY - Create the activity with subject, type, due date [Required]
  3. PIPEDRIVE_UPDATE_AN_ACTIVITY - Modify activity details or mark as done [Optional]
  4. PIPEDRIVE_GET_DETAILS_OF_AN_ACTIVITY - Retrieve activity record [Optional]
  5. PIPEDRIVE_GET_ALL_ACTIVITIES_ASSIGNED_TO_A_PARTICULAR_USER - List user's activities [Optional]

Key parameters:

  • subject: Activity title (required)
  • type: Activity type key string, e.g., "call", "meeting", "task", "email" (required)
  • due_date: Format YYYY-MM-DD
  • due_time: Format HH:MM
  • duration: Format HH:MM (e.g., "00:30" for 30 minutes)
  • deal_id / person_id / org_id: Link to related entities
  • done: 0 = not done, 1 = done

Pitfalls:

  • Both subject and type are required for PIPEDRIVE_ADD_AN_ACTIVITY
  • type must match an existing ActivityTypes key_string in the account
  • done is an integer (0 or 1), not a boolean
  • Response includes more_activities_scheduled_in_context in additional_data

4. Add and Manage Notes

When to use: User wants to attach notes to deals, persons, organizations, leads, or projects.

Tool sequence:

  1. PIPEDRIVE_SEARCH_PERSONS or PIPEDRIVE_GET_DETAILS_OF_A_DEAL - Resolve entity ID [Prerequisite]
  2. PIPEDRIVE_ADD_A_NOTE - Create note with HTML content linked to an entity [Required]
  3. PIPEDRIVE_UPDATE_A_NOTE - Modify note content [Optional]
  4. PIPEDRIVE_GET_ALL_NOTES - List notes filtered by entity [Optional]
  5. PIPEDRIVE_GET_ALL_COMMENTS_FOR_A_NOTE - Retrieve comments on a note [Optional]

Key parameters:

  • content: Note body in HTML format (required)
  • deal_id / person_id / org_id / lead_id / project_id: At least one entity link required
  • pinned_to_deal_flag / pinned_to_person_flag: Filter pinned notes when listing

Pitfalls:

  • content is required and supports HTML; plain text works but is sanitized server-side
  • At least one of deal_id, person_id, org_id, lead_id, or project_id must be provided
  • PIPEDRIVE_GET_ALL_NOTES returns notes across all entities by default; filter with entity ID params

5. Query Pipelines and Stages

When to use: User wants to view sales pipelines, stages, or deals within a pipeline/stage.

Tool sequence:

  1. PIPEDRIVE_GET_ALL_PIPELINES - List all pipelines and their IDs [Required]
  2. PIPEDRIVE_GET_ONE_PIPELINE - Get details and deal summary for a specific pipeline [Optional]
  3. PIPEDRIVE_GET_ALL_STAGES - List all stages, optionally filtered by pipeline [Required]
  4. PIPEDRIVE_GET_ONE_STAGE - Get details for a specific stage [Optional]
  5. PIPEDRIVE_GET_DEALS_IN_A_PIPELINE - List all deals across stages in a pipeline [Optional]
  6. PIPEDRIVE_GET_DEALS_IN_A_STAGE - List deals in a specific stage [Optional]

Key parameters:

  • id: Pipeline or stage ID (required for single-item endpoints)
  • pipeline_id: Filter stages by pipeline
  • totals_convert_currency: 3-letter currency code or "default_currency" for converted totals
  • get_summary: Set to 1 for deal summary in pipeline responses

Pitfalls:

  • PIPEDRIVE_GET_ALL_PIPELINES takes no parameters; returns all pipelines
  • PIPEDRIVE_GET_ALL_STAGES returns stages for ALL pipelines unless pipeline_id is specified
  • Deal counts in pipeline summaries use per_stages_converted only when totals_convert_currency is set

Common Patterns

ID Resolution

Always resolve display names to numeric IDs before operations:

  • Organization name -> org_id: PIPEDRIVE_SEARCH_ORGANIZATIONS with term param
  • Person name -> person_id: PIPEDRIVE_SEARCH_PERSONS with term param
  • Pipeline name -> pipeline_id: PIPEDRIVE_GET_ALL_PIPELINES then match by name
  • Stage name -> stage_id: PIPEDRIVE_GET_ALL_STAGES with pipeline_id then match by name

Pagination

Most list endpoints use offset-based pagination:

  • Use start (offset) and limit (page size) parameters
  • Check additional_data.pagination.more_items_in_collection to know if more pages exist
  • Use additional_data.pagination.next_start as the start value for the next page
  • Default limit is ~500 for some endpoints; set explicitly for predictable paging

Known Pitfalls

ID Formats

  • All entity IDs (deal, person, org, activity, pipeline, stage) are numeric integers
  • Lead IDs are UUID strings, not integers
  • Custom field keys are long alphanumeric hashes (e.g., "a1b2c3d4e5f6...")

Rate Limits

  • Pipedrive enforces per-company API rate limits; bulk operations should be paced
  • PIPEDRIVE_GET_ALL_PERSONS and PIPEDRIVE_GET_ALL_ORGANIZATIONS can return large datasets; always paginate

Parameter Quirks

  • Email and phone on persons are arrays of objects, not plain strings
  • visible_to is numeric: 1 = owner only, 3 = entire company, 5 = specific groups
  • done on activities is integer 0/1, not boolean true/false
  • Organization creation may auto-merge duplicates silently; check didMerge in response
  • PIPEDRIVE_SEARCH_PERSONS requires minimum 2 characters and does not support wildcards

Response Structure

  • Custom fields appear as hash keys in responses; map them via the respective Fields endpoints
  • Responses often nest data under response.data.data in wrapped executions
  • Search results are under response.data.items, not top-level

Quick Reference

TaskTool SlugKey Params
Create dealPIPEDRIVE_ADD_A_DEALtitle, value, org_id, stage_id
Update dealPIPEDRIVE_UPDATE_A_DEALid, status, value, stage_id
Get deal detailsPIPEDRIVE_GET_DETAILS_OF_A_DEALid
Search personsPIPEDRIVE_SEARCH_PERSONSterm, fields
Add personPIPEDRIVE_ADD_A_PERSONname, email, phone, org_id
Update personPIPEDRIVE_UPDATE_A_PERSONid, name, email
Get person detailsPIPEDRIVE_GET_DETAILS_OF_A_PERSONid
List all personsPIPEDRIVE_GET_ALL_PERSONSstart, limit, filter_id
Search organizationsPIPEDRIVE_SEARCH_ORGANIZATIONSterm, fields
Add organizationPIPEDRIVE_ADD_AN_ORGANIZATIONname, visible_to
Update organizationPIPEDRIVE_UPDATE_AN_ORGANIZATIONid, name, address
Get org detailsPIPEDRIVE_GET_DETAILS_OF_AN_ORGANIZATIONid
Add activityPIPEDRIVE_ADD_AN_ACTIVITYsubject, type, due_date, deal_id
Update activityPIPEDRIVE_UPDATE_AN_ACTIVITYid, done, due_date
Get activity detailsPIPEDRIVE_GET_DETAILS_OF_AN_ACTIVITYid
List user activitiesPIPEDRIVE_GET_ALL_ACTIVITIES_ASSIGNED_TO_A_PARTICULAR_USERuser_id, start, limit
Add notePIPEDRIVE_ADD_A_NOTEcontent, deal_id or person_id
List notesPIPEDRIVE_GET_ALL_NOTESdeal_id, person_id, start, limit
List pipelinesPIPEDRIVE_GET_ALL_PIPELINES(none)
Get pipeline detailsPIPEDRIVE_GET_ONE_PIPELINEid
List stagesPIPEDRIVE_GET_ALL_STAGESpipeline_id
Deals in pipelinePIPEDRIVE_GET_DEALS_IN_A_PIPELINEid, stage_id
Deals in stagePIPEDRIVE_GET_DEALS_IN_A_STAGEid, start, limit
Add product to dealPIPEDRIVE_ADD_A_PRODUCT_TO_A_DEALid, product_id, item_price

Powered by Composio

评价

还没有评价。欢迎分享你的使用感受。

评论

无需登录即可发表。欢迎分享使用感受、提问,或说明在哪些环境下无法运行。

还没有评论。来写第一条吧。

提交后会立即公开。请勿填写邮箱、电话等个人信息。URL 最多 2 个。

0 / 1000 字

此表单受 reCAPTCHA 保护,适用 Google 的隐私权政策和服务条款。

同一仓库的技能

Handle ActiveCampaign contact and marketing tasks through Rube MCP, a connection for agent tools. Find or create contacts, manage tags and subscriptions, enroll contacts in existing automated workflows, and create follow-up tasks.

中文简介正在准备中。当前显示英语简介。

  • Finding or registering contacts
  • Adding or removing contact tags
  • Subscribing contacts to lists or unsubscribing them
davepoon/buildwithclaude36662026年10月20日 更新

Research competitors’ live Meta and Google ads to identify opening hooks, offers, calls to action, and landing pages. Receive a sourced creative brief and up to three test ideas adapted to your product.

中文简介正在准备中。当前显示英语简介。

  • Find angles for new ad copy
  • Plan creative tests using competitor ads
  • Choose a landing page message
davepoon/buildwithclaude36662026年10月20日 更新

Analytics your AI agent can actually use. Track, analyze, run A/B experiments, and optimize across all your projects via CLI. Includes a growth playbook so your agent knows HOW to grow, not just what to track.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36662026年10月20日 更新

Teaches when to recall from long-term memory before acting and when to save durable decisions, corrections and failures afterwards. Use when a memory tool or MCP memory server is connected but the agent is not using it consistently, when the user complains that the assistant forgets preferences, conventions or past decisions between sessions, or when setting up persistent memory for a project. Works with any memory backend: a folder of Markdown files, a local MCP server, or a managed service.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36662026年10月20日 更新

Audit local AI coding-agent sessions with agenttrace for cost, tokens, tool failures, latency, anomalies, health, diffs, and CI gates.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36662026年10月20日 更新

Checks which private files (.env, keys, secrets) Claude Code and its subagents reached in this project, whether a value came into an agent's context, and which key to change, by running the open source agentwhy CLI over the session records Claude Code already keeps on disk. Use when the user asks whether the agent or a subagent read their .env or other secrets, which private files a session touched, whether a key leaked into the conversation, or which key to rotate. Runs locally and never prints a secret value.

中文简介正在准备中。当前显示原文说明。

davepoon/buildwithclaude36662026年10月20日 更新

查看 davepoon 的全部技能

报告此技能的问题