Editorial profile
Amplitude's official plugin layers reusable skills over its hosted analytics server. It covers account health, charts, dashboards, experiments, feedback, briefings, opportunity discovery, and instrumentation planning.
Where Amplitude fits
The useful question is not whether this agent plugin can be installed. It is whether its scope matches a named workflow, keeps the authoritative system clear, and gives a reviewer enough evidence to trust the result.
- Weekly product briefs
- Experiment and account-health analysis
- Analytics instrumentation planning
Documented capabilities
These are the practical capabilities described by the current public source. Confirm the exact tool surface, account limits, and enabled permissions in the client you plan to use.
- Create and analyze charts and dashboards
- Review experiments, feedback, and account health
- Plan and audit analytics instrumentation
Setup outline
Treat setup as a small integration project. Use a dedicated test identity, begin with the narrowest access available, and record who owns upgrades and credential revocation.
- Install amplitude@claude-plugins-official
- Connect the Amplitude MCP server
- Authenticate to the correct organization and verify project roles
What to check before adoption
Product documentation normally shows the happy path. The items below are the constraints or open questions most likely to affect a business rollout.
- Metric definitions, event names, identities, and cohorts must be verified.
- Creating content or changing instrumentation can affect shared analytics work.
Botfinder's take
Amplitude has the advantage of first-party provenance: the publisher controls the underlying product and its integration surface. That reduces one layer of ambiguity, but it does not remove the need to test permissions, failure handling, output quality, and the complete data path.
Start with a read or draft workflow where a person can compare the result with the source system. Add mutation only after the team can explain approvals, duplicate protection, partial failures, and recovery without relying on the model to infer whether a write succeeded.