agent-setup-maintenance

Name: agent-setup-maintenance
Availability: InStock
Author: Langfuse

Maintains Langfuse's shared agent configuration files and generated provider shims.

34.0k|3.7k|Updated May 18, 2023

Search Tags:#pnpm #langfuse #Monorepo #Workflow #developer tooling #Agent configuration #shim generation

Ready to run

npx skills add https://github.com/langfuse/langfuse --skill agent-setup-maintenance

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: agent-setup-maintenance
Source: https://github.com/langfuse/langfuse/tree/main/.agents/skills/agent-setup-maintenance
Command: npx skills add https://github.com/langfuse/langfuse --skill agent-setup-maintenance

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Editing Langfuse's repo-owned agent setup under .agents/ risks breaking generated provider shims, install-time sync, and discovery symlinks if done without a consistent workflow.

Core Features & Use Cases

Canonical Editing Workflow: Directs changes to the canonical .agents/ files instead of generated provider outputs like .claude/, .cursor/, or .codex/.
Install-Time Safety: Ensures postinstall agent sync logic skips cleanly in pruned Docker build contexts where repo files are absent.
Required Verification: Enforces running pnpm run agents:sync and pnpm run agents:check after any shared setup change.
Use Case: When adding a new shared skill or changing shim generation, follow this workflow to update the source of truth, keep AGENTS.md symlinks intact, and verify sync behavior in the same PR.

Quick Start

Ask the agent to update the shared agent setup in .agents/ and run the required sync and check verification commands.

Frequently Asked Questions about agent-setup-maintenance

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I update shared agent configuration in the Langfuse repo?▼

Edit the canonical files under `.agents/` rather than generated provider outputs, then run `pnpm run agents:sync` and `pnpm run agents:check` to regenerate and verify shims. Update README and contributor docs in the same PR when behavior changes materially.

How do I add a new shared skill to the Langfuse agent setup?▼

Create the skill with `SKILL.md` as its entrypoint, following the skill-creator skill and the repo-specific checks in this workflow. Do not add skill-by-skill links to the root `AGENTS.md`; keep discovery through the shared setup model.

Should I edit .claude or .cursor directories directly in Langfuse?▼

No, tool-specific directories like `.claude/`, `.cursor/`, `.codex/`, and `.vscode/` are generated discovery surfaces. Edit the canonical `.agents/` sources instead, unless a provider requires a fixed discovery path that the shared model cannot express.

Why does agent sync fail during Docker builds with pnpm install?▼

Turbo's pruned install stage can run root `postinstall` before `scripts/` and `.agents/` exist in the image. Install-time agent setup logic must detect missing repo-owned files and skip cleanly in those pruned contexts.

What verification is required after changing Langfuse agent setup?▼

Run `pnpm run agents:sync` and `pnpm run agents:check` after any shared setup change. Additionally run `pnpm run postinstall` when install-time behavior changes, plus targeted tests for any modified scripts.