For AI agents: the docs index is at /llms.txt. Every page has a markdown version: add .md to the URL.

Agent Coach
Security scan: pass (whole repo scanned)MCP serverMSection: Go-to-marketGmLicense: MITMIT

Amplitude AI Plugin (official MCP + skills)

Official Amplitude plugin for AI coding tools: connects Amplitude's MCP (OAuth to your Amplitude org) and ships skills to create and analyze charts and dashboards, design and read A/B experiments, triage session replays and errors, analyze AI-agent sessions, and plan analytics instrumentation from a PR diff.

Scan
Repo scan pass
GitHub stars
45
Last updated
2026-10-16
License
MIT
Cost
freemium
Why it helps

Lets a venture-operator agent instrument its product's events and then answer 'did this launch or experiment move the metric?' from real Amplitude data, so growth decisions are measured instead of guessed.

Ask your owner first. cost: freemium; needs a key or account; installs a mcp

  • Authorizes the agent against your Amplitude org via OAuth: connect a project you are fine letting the agent read and edit charts in.

Paste to your agent

Read https://upskill.coach/mcp/amplitude-mcp-plugin.md. If the security scan passed, ask me before installing, then install it with the command under Install and tell me what you did.

Install

claude plugin install amplitude (Claude Code), or codex plugin marketplace add amplitude/mcp-marketplace, or gemini extensions install https://github.com/amplitude/mcp-marketplace; then authenticate to your Amplitude org when prompted

Community

– views– downloads– comments

Helps with

Go-to-marketDeveloperMarketingOps automationEvals & qualityVenture operatorMeasurementQuality verification
Repo security-scanned on 2026-11-18 at commit 658f88b994d8: pass. The verdict covers the whole repo (amplitude/mcp-marketplace); this mcp wasn't scanned separately.
Scan notes
Automated scan: pass (no medium+ findings) at commit 658f88b994d8 on 2026-11-18.
Source
github.com/amplitude/mcp-marketplace
Parent repo
amplitude/mcp-marketplace

Static scan at a pinned commit (prompt-injection, malware and exfiltration rules plus semgrep, gitleaks, osv-scanner); flagged lines reviewed by hand. A scan lowers risk; it doesn't prove safety.

Alternatives

Compare all

For agents: markdown, JSON.

Comments

Comments are from visitors, not reviewed by Upskill, and never shown to agents.