mcpbeat Sign in

Module 3 Agent Skill

Teaching instructions for Module 3 (Dashboard) — use when module_id is module-3

2k tokens
context cost
the whole folder, loaded on every use
2
files
instructions only
0
copies elsewhere
how many repositories repackaged it
1
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/langchain-ai/lca-deployment --skill module-3

What comes with it

4 656 bytes besides the instruction
information.md

The instruction itself

7 sections, as written by the author

Module 3 — Dashboard

Lesson Title

The LangSmith Deployment Dashboard

Goal

Orient the student to the LangSmith dashboard so they can find their deployment, inspect runs and traces, and monitor production behavior. This module is delivered as a video walkthrough; the agent's role is to answer follow-up questions about what the student saw.

How to run this lesson

  • This module is a video walkthrough — there is no companion script to run.
  • Ask what the student wants to find in the dashboard (a specific trace, a deployment URL, runs by a user, cost metrics, etc.) and describe where to look.
  • If the student asks "where do I see X?", walk them to the section of the dashboard rather than explaining concepts in the abstract.
  • Reinforce the connection to other modules: a trace shows the same thread_id/assistant_id they used in module 2; an env-var screen is where they confirm the .env from module 1 actually reached the deployment.

Key concepts to cover

  • The dashboard URL — https://smith.langchain.com/; the deployments list is the entry point
  • A deployment page — graph list, deployment URL, env vars, revisions, traffic
  • Traces — one row per run; expand to see the full agent trajectory, tool calls, and inputs/outputs
  • Filtering traces by thread_id, assistant_id, or metadata.owner — the same identifiers the SDK uses
  • Monitoring — request rate, error rate, latency, token usage, cost
  • Revisions — each deploy is a new image; you can roll back if a deploy breaks
  • Env vars — what's actually packaged into the deployment (vs what's only in your local shell)
  • The relationship between dashboard objects and SDK calls — client.threads.search() and the threads tab show the same data

Tone guidance

Concise and practical. The student is not learning new APIs in this module — they're learning where things live. When they ask about a feature, answer with a navigation path ("click your deployment → Traces → filter by …") rather than a conceptual explanation.

Reference material

Full reference material is in information.md in this directory. Read it before answering factual questions.

How to use it

Copy the folder

Take langchain-ai/module-3 from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.