Bidirectional, idempotent sync of contacts/deals between a local SQLite source-of-truth and CRM APIs (HubSpot/Pipedrive/Salesforce) via one config.
Part of the Cognis Neural Suite.
pip install cognis-crmsync
crmsync scan . # → prioritized findings in seconds-
Install the CLI (Python 3.9+):
pip install crmsync # or: pip install . from a checkout -
Detect drift — the
diffsubcommand compares a versioned CRM export (CSV/TSV/JSON) against a local SQLite store and exits1when they differ:crmsync diff contacts.csv --db crm.db --table contacts --key email
-
Sync the DB to match the export (idempotent) with
apply:crmsync apply contacts.csv --db crm.db --table contacts --key email
Add
--no-deleteto keep DB rows that are absent from the export. -
Read drift programmatically with the global
--format jsonflag (it precedes the subcommand):crmsync --format json diff deals.json --db crm.db --key deal_id | jq .
-
Gate CI on CRM consistency — the job fails (exit 1) whenever the export and DB drift apart:
crmsync diff contacts.csv --db crm.db --key email || echo "CRM export drifted from DB"
- Why crmsync? · Features · Quick start · Example · Architecture · AI stack · How it compares · Integrations · Install anywhere · Related · Contributing
A single MCP-native binary that makes your CRM a replica of a versioned local file — run it in CI to detect drift and reconcile without a $20k iPaaS contract.
crmsync is single-purpose, scriptable, and self-hostable: point it at a target, get prioritized results in the format your workflow already speaks (table · JSON · SARIF), gate CI on it, and let agents drive it over MCP.
- ✅ Fingerprint
- ✅ Load Export
- ✅ Ensure Schema
- ✅ Load Db State
- ✅ Diff Records
- ✅ Apply Plan
- ✅ Runs on Linux/macOS/Windows · Docker · devcontainer
- ✅ Ports in Python, JavaScript, Go, and Rust (
ports/)
pip install cognis-crmsync
crmsync --version
crmsync scan . # scan current project
crmsync scan . --format json # machine-readable
crmsync scan . --fail-on high # CI gate (non-zero exit)$ crmsync scan .
[HIGH ] CRM-001 example finding (./src/app.py)
[MEDIUM ] CRM-002 another signal (./config.yaml)
2 findings · risk score 5 · 38ms
flowchart LR
IN[input] --> P[crmsync<br/>analyze + score]
P --> OUT[report]
crmsync is interoperable with every popular way of using AI:
- MCP server —
crmsync mcp(Claude Desktop, Cursor, Cognis.Studio, uncensored-fleet) - OpenAI-compatible / JSON — pipe
crmsync scan . --format jsoninto any agent or LLM - LangChain · CrewAI · AutoGen · LlamaIndex — wrap the CLI/JSON as a tool in one line
- CI / scripts — exit codes + SARIF for non-AI pipelines
| Cognis crmsync | Airbyte | |
|---|---|---|
| Self-hostable, no account | ✅ | varies |
| Single command, zero config | ✅ | |
| JSON + SARIF for CI | ✅ | varies |
| MCP-native (AI agents) | ✅ | ❌ |
| Polyglot ports (JS/Go/Rust) | ✅ | ❌ |
| Open license | ✅ COCL | varies |
Built in the spirit of Airbyte/Singer taps, in the spirit of Grouparoo, re-framed the Cognis way. Missing a credit? Open a PR.
Pipes into your stack: SARIF for code-scanning, JSON for anything, an MCP server (crmsync mcp) for AI agents, and a webhook forwarder for SIEM/Slack/Jira. See docs/INTEGRATIONS.md.
pip install "git+https://github.com/cognis-digital/crmsync.git" # pip (works today)
pipx install "git+https://github.com/cognis-digital/crmsync.git" # isolated CLI
uv tool install "git+https://github.com/cognis-digital/crmsync.git" # uv
pip install cognis-crmsync # PyPI (when published)
docker run --rm ghcr.io/cognis-digital/crmsync:latest --help # Docker
brew install cognis-digital/tap/crmsync # Homebrew tap
curl -fsSL https://raw.githubusercontent.com/cognis-digital/crmsync/main/install.sh | sh| Linux | macOS | Windows | Docker | Cloud |
|---|---|---|---|---|
scripts/setup-linux.sh |
scripts/setup-macos.sh |
scripts/setup-windows.ps1 |
docker run ghcr.io/cognis-digital/crmsync |
DEPLOY.md (AWS/Azure/GCP/k8s) |
warmline— Score and rank inbound/outbound leads from a YAML rulebook, emitting a ranked queue as JSON/CSV for your SDRs and CI gates.coldforge— Render personalized cold-outreach sequences from Markdown templates + a contacts CSV, with spam-score linting and per-send dry-run preview.pactgen— Generate branded sales proposals and SOWs from a YAML scope file + pricing table into PDF/HTML, with a deterministic line-item math check.dripcheck— Lint email sequences and drip campaigns for deliverability: SPF/DKIM/DMARC, link health, unsubscribe presence, and CAN-SPAM/GDPR compliance.dealflow— Model your sales pipeline as a YAML state machine and compute conversion rates, stage velocity, and weighted forecast straight from CRM exports.introbot— Find warm-intro paths through your team's combined network graph and draft double-opt-in intro requests from a single contacts manifest.
Explore the suite → 🗂️ all 170+ tools · ⭐ awesome-cognis · 🔗 cognis-sources · 🤖 uncensored-fleet · 🧠 engram
PRs, new rules, and demo scenarios are welcome under the collaboration-pull model — see CONTRIBUTING.md and SECURITY.md.
{} composes with the 300+ tool Cognis suite — JSON in/out and a shared
OpenAI-compatible /v1 backbone. See INTEROP.md for the
suite map, composition patterns, and reference stacks.
Source-available under the Cognis Open Collaboration License (COCL) v1.0 — free for personal, internal-evaluation, research, and educational use; commercial / production use requires a license (licensing@cognis.digital). See LICENSE.