amplitude
This skill queries read-only product analytics from Amplitude, including active users, events, funnels, retention, experiments, cohorts, session replays, and…
Identity proven by GitHub's verified organisation badge (getsentry)
- Agents
- Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode, claude.ai and API
- License
- Apache-2.0 (permissive)
- Pin
- 1745a29pinned Oct 9, 2026
- Context cost
- about 1,100 tokenswhen the skill loads
- GitHub stars
- 367on getsentry/junior
- Scripts
- None found
What it does
“Query read-only Amplitude product analytics. Use when users ask about active users, product usage, events, segmentation, funnels, conversion, retention, charts, dashboards, experiments, cohorts, taxonomy, session replay, feature flags, guides, surveys, feedback, agent analytics, or individual user activity in Amplitude. Do not use for changing Amplitude charts, dashboards, experiments, cohorts, flags, taxonomy, or project configuration.”
Generated summary from the skill's own text
This skill queries read-only product analytics from Amplitude, including active users, events, funnels, retention, experiments, cohorts, session replays, and feature flags. The agent connects through Amplitude's MCP server to retrieve metrics and analysis results without modifying any Amplitude resources. It handles metric definitions, time ranges, and result interpretation for product usage investigations.
When to use it
- Analyzing user engagement metrics like DAU, WAU, or MAU over specific periods
- Investigating event flows, funnel conversion rates, and user retention patterns
- Retrieving saved charts, dashboards, or experiment results by identifier
- Examining feature flag configurations or session replay details for specific users
When not to
- Modifying Amplitude charts, dashboards, experiments, cohorts, or project settings
- Creating new analyses when the user lacks Amplitude access or authorization
- Querying when explicit event or property names cannot be preserved exactly
Install
These commands install from the publisher's repository and need no account.
skills CLI
npx skills add getsentry/junior --skill amplitudeInstalls from the repository's default branch, which may have changed since the commit we checked.
GitHub CLI 2.90 or later
gh skill install getsentry/junior amplitude@1745a298365f0f001db0b3eec6de8a8347e0f6d9Installs the commit we checked (1745a29).
Copy by hand
Download the folder at 1745a29 and put it in .claude/skills/amplitude for Claude Code and in .agents/skills/amplitude for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode.
Claude Code plugin
/plugin install sentry@claude-plugins-officialThe publisher's plugin in Claude's official marketplace. It may hold more than this skill, and we have not pinned it.
Claude Code plugin
/plugin install sentry-cli@claude-plugins-officialThe publisher's plugin in Claude's official marketplace. It may hold more than this skill, and we have not pinned it.
Copy for your agent
A prompt for any coding agent. It installs this skill from GitHub at the pinned commit, copies the files as they are, and runs nothing unless you ask.
Prompt
Install the agent skill "amplitude" by Sentry. Source, pinned to commit 1745a298365f0f001db0b3eec6de8a8347e0f6d9: https://github.com/getsentry/junior/tree/1745a298365f0f001db0b3eec6de8a8347e0f6d9/packages/junior-amplitude/skills/amplitude 1. Download that folder at exactly this commit. Do not use the default branch. 2. Copy every file unchanged, keeping the folder structure, into .claude/skills/amplitude/ for Claude Code and .agents/skills/amplitude/ for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. Create the folders if they do not exist. 3. Do not edit the files, and do not run any script in the folder unless I ask you to. 4. When you are done, list the files you copied and where you put them. Listing and checks: https://www.agentnative.dev/skills/sentry/amplitude
SKILL.md at the pin
--- name: amplitude description: Query read-only Amplitude product analytics. Use when users ask about active users, product usage, events, segmentation, funnels, conversion, retention, charts, dashboards, experiments, cohorts, taxonomy, session replay, feature flags, guides, surveys, feedback, agent analytics, or individual user activity in Amplitude. Do not use for changing Amplitude charts, dashboards, experiments, cohorts, flags, taxonomy, or project configuration. --- # Amplitude Analytics Use this skill for read-only product analytics through Amplitude's hosted MCP server. ## Workflow 1. Define the question before querying: - Identify the metric or analysis type: active users, event segmentation, funnel, retention, chart, dashboard, experiment, cohort, taxonomy, session replay, feature flag, guide or survey, feedback, agent analytics, or user activity. - Preserve explicit event names, property names, chart IDs, dashboard IDs, experiment IDs, cohort IDs, and user IDs exactly. - Use the user's explicit time range. For relative ranges, state the concrete dates in the answer. - If the request has no time range, use the shortest conventional range that answers it and state the assumption. 2. Discover the current Amplitude tool: - Inspect the available Amplitude MCP tools for the required analysis. - Use the exact live tool name and schema. Do not guess Amplitude MCP tool names or arguments. - Prefer direct chart, dashboard, experiment, cohort, or user lookup when the user provides its identifier. - If project or organization scope is ambiguous, inspect accessible scope through the read-only catalog and ask one focused question only when multiple plausible targets remain. 3. Query narrowly: - Request only the events, properties, segments, metrics, and dates needed for the answer. - Prefer one complete query over several overlapping queries. - For comparisons, keep metric definitions and windows consistent. - Stop once the requested result is supported; do not enumerate unrelated project metadata. 4. Report the result: - Lead with the metric, trend, conversion, retention, or experiment finding. - Include the date range, project or scope, filters, and metric definition needed to interpret it. - Distinguish unique users, event totals, sessions, and percentages explicitly. - Include Amplitude links returned by the MCP server when available. - State when data is empty, delayed, inaccessible, or insufficient rather than inferring a value. ## Common Analyses - **DAU, WAU, MAU:** use active-user or event-segmentation tools with the requested interval and clarify whether the result is unique users or event totals. - **Event segmentation:** preserve the exact event and property filters; report the aggregation and interval. - **Funnels:** preserve ordered steps, conversion window, exclusions, and segment filters; report overall conversion and material drop-off steps. - **Retention:** report the starting event, returning event, cohort interval, and retention horizon. - **Saved charts and dashboards:** fetch by ID when available and summarize the returned definition and results without recreating the analysis from memory. - **Experiments:** report status, variants, exposure window, primary metric, and result confidence exactly as returned. - **Session replay:** narrow by time, user, or event before retrieving replay details; summarize only the interactions needed to answer the question. - **Feature flags:** report definitions, variants, and configuration without exposing deployment API keys or other credential fields. - **Guides, surveys, and feedback:** preserve source and date filters, distinguish processed themes from raw comments, and avoid returning unrelated customer text. - **Agent analytics:** distinguish sessions, spans, conversations, quality metrics, cost, and latency; return transcript content only when the user explicitly requests it. - **User activity:** use an explicit user identifier when possible and avoid exposing unrelated profile properties or event history. ## Guardrails - Read-only only. The plugin exposes a fixed allowlist of Amplitude search, retrieval, and query tools. - Do not create, update, archive, delete, launch, stop, or otherwise modify Amplitude resources. - If the user asks for a change, explain that this plugin is read-only and offer to inspect the current state or describe the change they would need to make. - Do not weaken or bypass the plugin's tool filter, including by calling Amplitude APIs through shell commands. - Treat event and user properties as potentially sensitive. Return only fields needed to answer the request. - Never expose API keys, credential fields, or unrelated conversation and feedback content returned by read tools. - Do not fabricate event names, property values, metric definitions, experiment conclusions, or missing data. - On authorization failure, let Junior present Amplitude's OAuth flow. On permission failure, report that the connected Amplitude account cannot access the requested scope.
From getsentry/junior at 1745a29, shown as plain text under its license: Apache-2.0 (permissive). The license text is in LICENSE.
View SKILL.md on GitHub at 1745a29Files
- SKILL.md5.0 KB · 68c3e94b57a1
- SOURCES.md3.8 KB · cba4705d9d17
- SPEC.md2.3 KB · 6a35b4e67b06
3 files, 11 KB in total. Each file has a sha256 in the pin's manifest; the first 12 characters are shown.
Skill folder on GitHub at 1745a29More from Sentry
claude-settings-audit
This skill analyzes a repository to detect its tech stack, build tools, and service integrations, then generates recommended Claude Code settings.json…
Claude Code · about 2.9K tokens · pinned 6 days ago
commit
This skill enables the agent to create conventional commit messages following Sentry's style guide.
Claude Code, Codex, Cursor +4 · about 430 tokens · pinned 6 days ago
django-access-review
This skill guides agents through investigating Django and Django REST Framework code for IDOR and access control vulnerabilities.
Claude Code, Codex, Cursor +4 · about 2.5K tokens · pinned 6 days ago
django-perf-review
This skill audits Django codebases for performance problems by validating issues through code tracing before reporting.
Claude Code, Codex, Cursor +4 · about 2.8K tokens · pinned 6 days ago
iterate-pr
This skill automates PR iteration by fetching CI check statuses and review feedback, then fixing actionable failures and high/medium priority comments until…
Contains scripts
Claude Code · about 1.2K tokens · pinned 6 days ago
pr-writer
This skill helps agents write clear, reviewer-focused PR titles and descriptions by inspecting the branch, diff, and existing PR state, then crafting concise…
Claude Code, Codex, Cursor +4 · about 1.3K tokens · pinned 6 days ago
Related in Data and databases
data-table-manager
This skill manages n8n Data Tables through creation, import, querying, and mutation operations.
by n8nOfficial publisher
Claude Code, Codex, Cursor +3 · about 2.1K tokens · pinned 6 days ago
algolia-automation
This skill automates Algolia search and indexing operations through the Rube MCP interface.
by ComposioOfficial publisher
Claude Code, Codex, Cursor +3 · about 690 tokens · pinned 6 days ago
expo-sqlite
This skill enables SQLite database operations in Expo and React Native apps for structured data storage, offline queues, and persistent caches.
by ExpoOfficial publisher
Claude Code, Codex, Cursor +4 · about 2.6K tokens · pinned 6 days ago
prisma-8
This skill helps agents work with Prisma 8, a contract-first data layer for TypeScript projects.
by PrismaOfficial publisher
Contains scripts
Claude Code, Codex, Cursor +3 · about 3.5K tokens · pinned 6 days ago
analyzing-experiment-session-replays
This skill enables agents to analyze session recordings from experiment variants to identify behavioral differences between control and test groups.
by PostHogOfficial publisher
Claude Code, Codex, Cursor +4 · about 2.3K tokens · pinned 6 days ago
auditing-warehouse-view-health
Audits materialized views in a PostHog data warehouse to identify failed views and flag unused ones that waste compute resources.
by PostHogOfficial publisher
Claude Code, Codex, Cursor +4 · about 1.4K tokens · pinned 6 days ago