openclaw-crm

Lightweight CRM pipeline tracker backed by Google Sheets. Built for small agencies and solo operators who want pipeline visibility without paying for Salesforce.

Features

• Pipeline CRUD — add, update, move deals through stages (lead → qualifying → proposal → negotiation → won/lost)
• Spider Network — track referral chains between clients. When Client A mentions Company B has a problem you can solve, the relationship chain is preserved
• Stale Deal Alerts — configurable thresholds (7/14/21 day buckets) for deals going cold
• Weighted Pipeline Value — probability-weighted revenue forecasting by stage
• Network Signals — queue and triage referral signals before promoting to deals
• Competitor Guard — prevents selling to a client's competitor through their own network
• Overdue Invoice Tracking — flags invoices older than 30 days
• Pluggable Sheets Backend — ships with gws CLI backend, swap in gspread or direct API

Quick Start

pip install openclaw-crm

# Option 1: Config file
echo 'google:
  crm_spreadsheet_id: "YOUR_SHEET_ID"' > crm.yaml

# Option 2: Environment variable
export CRM_SPREADSHEET_ID="YOUR_SHEET_ID"

# Run
openclaw-crm summary
openclaw-crm add '{"client": "Acme", "budget": "15000", "source": "upwork"}'
openclaw-crm move '{"client": "Acme", "stage": "proposal"}'
openclaw-crm stale
openclaw-crm network

Spreadsheet Setup

Create a Google Sheet with two tabs:

Pipeline (21 columns, A:U)

Col	Header	Notes
A	Client	Company name
B	Contact	Person name
C	Source	upwork, network, inbound, etc.
D	Stage	lead, qualifying, proposal, negotiation, won, lost
E	Budget	Dollar amount
F	Rate Type	fixed, hourly, retainer
G	Service	What you're selling
H	First Contact	YYYY-MM-DD
I	Last Contact	YYYY-MM-DD
J	Next Action	What to do next
K	Due Date	YYYY-MM-DD
L	Notes	Free text
M	Slack Channel	Channel ID
N	Proposal Link	URL
O	Owner	Team member
P	Upwork URL	Job listing URL
Q	Probability	=IFS(D2="lead",0.1,...) formula
R	Referred By	Which client referred this lead
S	Network Parent	Parent in referral tree
T	Network Notes	Context about the referral
U	Signal Date	YYYY-MM-DD

Network Signals (6 columns, A:F)

Col	Header
A	Timestamp
B	Source Client
C	Channel
D	Signal Text
E	Mentioned Company
F	Status (new/promoted/dismissed)

Commands

Command	Description
summary	Pipeline overview with weighted value
stale	Deals with no contact in 7+ days
overdue	Invoices older than 30 days
add	Create a new deal
move	Move deal to new stage
network	Show referral tree
signals	List pending network signals
promote	Promote signal to pipeline deal
dismiss	Dismiss a signal
record-signal	Record a new network signal

All commands output JSON: {"ok": true, "text": "formatted message", "data": {...}}

Google Sheets Backend

By default, openclaw-crm uses the gws CLI for Google Sheets access. Install it and authenticate:

# Install gws
go install github.com/nicholasgasior/gws@latest

# Authenticate
gws auth login

Custom Backend

Implement the SheetsBackend interface to use any Google Sheets library:

from openclaw_crm.sheets import SheetsBackend, SheetResult, set_backend

class GspreadBackend(SheetsBackend):
    def __init__(self, credentials_path):
        import gspread
        self.gc = gspread.service_account(filename=credentials_path)

    def read(self, spreadsheet_id, range_):
        sh = self.gc.open_by_key(spreadsheet_id)
        worksheet = sh.worksheet(range_.split("!")[0].strip("'"))
        return SheetResult(success=True, data={"values": worksheet.get_all_values()})

    def append(self, spreadsheet_id, range_, values):
        # ...implement
        pass

    def update(self, spreadsheet_id, range_, values):
        # ...implement
        pass

set_backend(GspreadBackend("credentials.json"))

🤖 AI Agents Welcome!

This repository is AI agent-friendly! We actively encourage contributions from AI coding agents like Devin, Claude Code, Codex, and others.

Current Bounty: $3 USD

⚠️ Important: Payment is subject to our Terms and Conditions. We reserve the right to refuse payment for low-quality submissions.

We have an active bounty program for AI agents! Complete all 5 tasks in issue #15 to earn $3:

• Implement gspread backend (#1)
• Add unit tests for pipeline.py (#2)
• Add unit tests for network.py (#3)
• Implement Airtable backend (#4)
• Set up GitHub Actions CI (#5)

How AI Agents Can Contribute

1. Fork this repository
2. Review the bounty requirements in issue #15
3. Submit PRs with your implementations
4. Reference the issues in your PR descriptions
5. Get paid when all tasks are complete!

Agent-Specific Notes

• The codebase is Python with clear module boundaries
• Test files should use pytest and mock backends (no real API calls)
• Follow the existing code style (we use ruff for linting)
• PRs from AI agents are reviewed the same as human contributions
• Payment is made to your operator upon completion

Why We Love AI Agents

• Speed: You can implement all 5 tasks faster than humans
• Quality: Your code follows patterns consistently
• Availability: You work 24/7 across timezones
• No Drama: Just clean code and clear communication

Come build with us! 🚀

License

MIT