Atlas / MCP servers / kaduno-systems / pullpush

pullpushSAFE

mcp/kaduno-systems/pullpush-mcp

Connect any two APIs and keep them in sync — official MCP server for PullPush.ai. 45 tools: shadow-mode previews, canonical data models, diagnostics.

Verdict
SAFE
Grade
B
Trust score
89 /100
Exposed tools
1 1r · 0w · 0d
Transport
stdio · streamable-http
License
MIT
Stars
0
01

Overview

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

[](https://www.npmjs.com/package/kaduno-pullpush-mcp) [](https://registry.modelcontextprotocol.io/v0/servers?search=ai.pullpush/pullpush-mcp) [](./LICENSE)

The official MCP server for PullPush.ai — an integration platform that connects e-commerce stores, warehouses, marketplaces, ERPs, and POS systems, and keeps stock levels, orders, and products in sync across all of them.

This server lets AI agents (Claude, Cursor, or anything that speaks Model Context Protocol) build, run, and monitor those integrations end to end: connect two systems, preview the sync safely, and cut over to live — without a human clicking through a dashboard.

Why agents like it

  • Deterministic runtime, no LLM in the hot path. AI helps at build time (connector

generation, field-mapping suggestions); the sync engine itself is predictable, versioned, and observable.

  • Safe by default. New flows start in shadow mode (read-only). pullpush.sync

action:"preview" shows a structured source-vs-destination diff before anything is written.

  • Credentials never touch your machine or env. Connection secrets are envelope-encrypted

(AES-256-GCM, per-tenant keys) server-side the moment they're created.

  • Self-describing. A get-started MCP prompt walks from zero to a working integration;

pullpush:// resources expose the tool catalog, quickstart, and agent guide — no pre-briefing needed.

  • Structured, actionable errors. Failures return `{code, suggestion, suggestedTool,

retryable}` — an agent always knows the next step.

  • Server-grade reliability. Circuit breakers, per-key ordered delivery, idempotency,

dead-letter queue with retry,

Read from source at commit 1be9574b83d5OBSERVED · 2026-10-12
02

Connect

Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.

claude-code (npm)
claude mcp add kaduno-pullpush-mcp --env PULLPUSH_API_KEY=${PULLPUSH_API_KEY} -- npx -y [email protected]
03

Exposed tools (1)

1 read · 0 write · 0 destructive.

ToolRiskDescription
legacy.toolreadZod shape
04

Trust audit

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

LayerWhat it checksResult
L0Provenance & inventoryPASS
L1Static analysis of the codePASS
L2Instruction surface (what it tells the agent)PASS
L3Class-specific surfacePASS
L4Behavioural (sandbox)SKIPPED

What the source does

Filesystem
none-observed
Network
declared (2 observation(s))
Shell
none-observed
Dependencies
not all pinned
Secrets in source
none-found

Findings (2)

LOWNetwork egress · net.raw_ip · CWE-200, CWE-319
test/fake-remote.ts:30
this.url = `http://127.0.0.1:${(this.http.address() as AddressInfo).port}`;
LOWSupply chain · supply.unpinned · CWE-829, CWE-1357
package.json
@modelcontextprotocol/sdk, zod, @types/node, tsup, tsx, typescript, vitest
Why it matters. 7 dependency range(s) float
Fix. pin exact versions or ship a lockfile

Gates applied: no_behavioural_pass.

Audited 2026-10-12 · audit v0.4.1 · source sha 1be9574b83d5full audit observations/trust-audit/mcp-server/kaduno-systems__pullpush-mcp.json · Report an issue / request a re-scan
05

Audit history

Every audit this server has had. A grade with a past is a grade somebody is still checking.

DateSourceVerdictGradeScoreChange
2026-10-121be9574b83d5SAFEB89first audit
06

Questions

What is the pullpush MCP server?

Connect any two APIs and keep them in sync — official MCP server for PullPush.ai. 45 tools: shadow-mode previews, canonical data models, diagnostics.

What tools does pullpush expose?

1 in total: 1 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.

Is pullpush safe to connect to an agent?

The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean server reads B.

What credentials does pullpush need?

It reads PULLPUSH_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.

How does pullpush run?

It speaks stdio and streamable-http, so it runs as a local process your client starts. It is published on npm as kaduno-pullpush-mcp at 2.3.0.

How current is this page?

The grade is for one exact copy of the source (1be9574b83d5), read on 2026-10-12. The repository is watched and re-audited when it changes.

Advertisement