langfuse

Prepare Langfuse weekly production reviews covering failures, fixes, open issues, and tracking gaps. Use for "what broke last week," production bugs, Datadog alerts or error patterns, incident.io activity, or pager load.

Configure and troubleshoot Turborepo monorepos. Use for turbo.json, task pipelines, dependsOn, caching, filters, affected packages, CI optimization, environment variables, package boundaries, or shared internal packages.

Use when writing or reviewing Storybook stories (`.stories.tsx`) for React components.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Codex's capabilities with specialized knowledge, workflows, or tool integrations.

Decide whether and how errors report to Sentry. Use when touching capture or error-handling paths in `web/**`, triaging Sentry noise, or changing Sentry settings.

Seed reproducible local Langfuse data in ClickHouse and Postgres. Use for complex traces, long sessions, v3/v4 events, bulk list data, or frontend rendering and performance tests; never use ad hoc scripts or raw inserts.

Review Langfuse changes for SSRF, tenant isolation, secret handling, unsafe redirects or uploads, RBAC drift, and client telemetry privacy. Use when a design or change accepts URLs or host fields, handles secrets or cross-tenant data, makes outbound requests, adds an integration, follows redirects, widens permissions, or can send customer-controlled UI data through analytics, monitoring, or session replay.

Refactor avoidable React useEffect usage in Langfuse frontend code. Use when adding, reviewing, or removing effects; initializing forms or local UI state from query data; synchronizing client and server state; moving mutations or async workflows out of components; cleaning every effect from a frontend submodule; or reviewing whether an effect has a valid external-system owner.

Guidelines for writing React components. Use this when creating a new react component.

Use this skill to clean up a React component. Only use this skill when instructed to do so by the user.

Product analytics with posthog. Use when adding a meaningful user action or feature in `web/**`, touching PostHog capture code, changing session replay or its privacy boundaries, or answering product-usage questions.

Upgrade pnpm workspace dependencies to target/latest versions: direct/transitive bumps, release-age checks, temporary overrides, minimumReleaseAgeExclude, lockfile/dedupe verification.

Answer "what should I do today" for a Langfuse maintainer, from the tracker rather than from memory: which projects you lead, which owe an update before the Monday engineering weekly, what shipped but is not finished, what waits on your decision, and what colleagues are working on that overlaps. Use on "what should I do today", "what's on my plate", "what should I work on next", "prep my Monday update", "what did I say last week", "who else is working on this", "who should review this", and whenever someone hands over a bare link — a ticket, a pull request, a Slack permalink — and expects you to take it from there.

Use Linear as the org's memory: reconstruct a feature's history before touching it, and leave the reasoning behind finished work in the ticket description so the next agent inherits it. Use when starting work on an existing feature, when splitting a large change, and always when wrapping up — "what happened to this screen before", "write the handover", "why is this code like this".

Use this skill after a bug or regression candidate has measured evidence to document them within linear.

The org policy for what an agent may write to Linear, and how it must be marked. Read this before creating a ticket, editing a description, or commenting on a ticket as an agent — "file this in Linear", "comment on the ticket", "create a subticket", "write the handover". Default those writes to a short human description; expand with research only when asked. Also covers the required Linear connection and what to do when there isn't one.

Use Langfuse's disposable per-PR previews at pr-N.preview.langfuse.com (synthetic data only). Use for preview access, failed deployments, test-data seeding, kubectl debugging, or waking sleeping previews.

Navigate Langfuse repositories, code areas, and agent skills. Use to locate code, choose the right repo or skill, search across the Langfuse organization, or orient before implementation, debugging, documentation, support, or operations.

Tune and review Langfuse autoscaling for web, web-iso, and web-ingestion. Use for Terraform scale settings, RPM targets, scaling bounds, task counts, cost/performance tradeoffs, or Datadog evidence in the infrastructure repo.

Read and record root causes in the Linear `incident-alert` knowledge base. Use before and after investigating a named Datadog monitor, incident.io alert or incident, or on-call page to find or record root causes.

Triage an engineer's work queue across Linear, Pylon, and GitHub. Use for assigned, urgent, waiting, or stale issues, support follow-ups, and PR reviews.

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Langfuse repo Git, GitHub, commit, branch, pull request, issue search, release, and production-promotion workflow. Use when staging, committing, pushing, opening PRs, choosing a Linear git branch name, handling Claude, Greptile, or Codex review comments, searching GitHub issues, or changing release/promotion behavior.

Architect large or state-heavy Langfuse frontend features. Use for virtualized lists, large tables, controllers, Zustand stores, row selection, high-frequency state, rendering performance, or before adding useEffect, useMemo, useCallback, or form defaults derived from loaded data.

Shared workflow for browser-based review of user-visible frontend changes in Langfuse. Use when a change affects UI behavior, layout, styling, navigation, or browser-visible regressions and should be checked with available browser automation before signoff.

Establish root cause by combining Datadog telemetry with the Langfuse repo. Use when investigating or triaging a user report, Linear or GitHub issue, incident, or pasted production error.

Research Langfuse production telemetry with reusable Datadog queries. Use for tenant or project activity, API usage, queue behavior, spans, logs, metrics, or ad hoc measurements across production regions; pair with debug-issue-with-datadog for root-cause analysis.

Human handoff, Linear branch names, reviewable (non-draft) PRs, the `cursor` GitHub label, Claude, Greptile, and Codex review comments, preview test steps, proof of work posted on the GitHub PR, and review-doubt notes for Cursor agents. Use when a Cursor Cloud or Cursor desktop agent implements a Linear issue, opens a GitHub PR, asks a human to test, posts screenshots or videos, or handles Claude, Greptile, or Codex code-review comments.

Design, implement, review, or harden Langfuse repo-owned autonomous agents. Use for LLM-powered GitHub Actions, scheduled or dispatched agents, agent-created PRs, prompts, allowlists, tokens, untrusted content, or self-updating instructions.

Review Langfuse code changes for correctness, regressions, and best practices.

MUST USE when reviewing ClickHouse schemas, queries, or configurations. Contains 28 rules that MUST be checked before providing recommendations. Always read relevant rule files and cite specific rules in responses.

Shared workflow for writing Langfuse changelog entries after a feature is complete. Use when a branch is ready for merge and a changelog entry or changelog draft is needed.

Build or review Langfuse backend code. Use for tRPC routers, public REST APIs, BullMQ processors, services, middleware, Prisma or ClickHouse access, OpenTelemetry, Zod, environment configuration, or backend tests.

Analyze Langfuse Cloud infrastructure cost structure using Metabase cost marts. Use when asked about cloud spend, AWS versus ClickHouse cost splits, cost drivers by provider/service/usage type/account, daily cost per tracing event, infra cost dashboards, or cost regressions visible in Metabase.

Shared workflow for editing Langfuse's repo-owned agent setup under `.agents/`. Use when changing AGENTS files, shared skills, `.agents/config.json`, generated shim behavior, provider discovery paths, or install-time agent sync.

Use when editing worker/src/constants/default-model-prices.json, packages/shared/src/server/llm/types.ts, pricing tiers, tokenizer IDs, or matchPattern regexes for OpenAI, Anthropic, Bedrock, Vertex, Azure, or Gemini model pricing.

Interact with Langfuse and access its documentation: tracing, monitoring, creating datasets, running experiments, and evaluating AI applications. Use when needing to (1) query or modify Langfuse data, (2) look up Langfuse documentation, concepts, integration guides, a feature or SDK usage, or (3) do any AI engineering task (AI observability, prompt engineering/management, evaluation and evaluator management, experimentation, dataset management, evaluation-driven CI/CD, feedback collection). Invoke it for tasks in this scope even when Langfuse is not configured or explicitly mentioned.