
Prompt Hub And Prompt Management
LangSmith · LangSmith
LLM Observability & Evaluation Platform — LangSmith (LangChain)
Evaluates LangSmith's Prompt Hub / Prompt Management across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's LLM Observability & Evaluation Platform eval coverage.
About LangSmith
LangSmith is LangChain's LLM observability and evaluation platform: tracing, datasets, evaluators (LLM-as-judge, code, and human), experiments, prompt management, and online monitoring used by AI teams to measure and improve LLM apps in production.
Employees
~200
Industry
LLM Observability
Headquarters
San Francisco, CA
Website
www.langchain.com/langsmithSample tests· showing 3 of 9
| # | Input | Expected behavior | Check |
|---|---|---|---|
| 01 | Operator pushes a new prompt 'support-triage' to the Prompt Hub with client.push_prompt. | push_prompt creates a new commit hash for the prompt. Record the returned commit hash in the deployment manifest (e.g., infra repo). Subsequent pull_prompt('support-triage:<commit_hash>') pins to that exact version. Pushing the same content twice is a no-op (same hash). | Pass / FailAi Platformhigh |
| 02 | Production deployment must pin to a specific prompt commit so a UI edit cannot silently change behavior. | Call client.pull_prompt('support-triage:<commit_hash>') at deploy time and cache the resulting prompt object in the process. Treat the commit hash as part of the deployment artifact. NEVER pull bare 'support-triage' in prod — that resolves to the latest commit and is mutable. | Pass / FailAi Platformcritical |
| 03 | Prompt is published as a runnable chain with include_model=true so the model binding lives alongside the prompt. | client.pull_prompt(name, include_model=True) returns a runnable that already binds the documented model+params. Use this only when the team wants model+prompt to version together. For ops control over model rollouts, keep model selection outside the prompt and pass include_model=False (default). | Pass / FailAi Platformmedium |
How this eval is graded
Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
Rubric criteria
- Langsmith
- Ai Platform
- Prompt Hub And Prompt Management
Recommended for
Works with
Related evals
Claude API
Evaluates Anthropic's Batch API across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Extended Thinking across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
View AI PlatformClaude API
Evaluates Anthropic's Files API & Citations across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's Foundation Model & API eval coverage.
ViewFrequently asked questions
What does the Prompt Hub And Prompt Management eval for LangSmith LangSmith test?+
Evaluates LangSmith's Prompt Hub / Prompt Management across 9 scenario-based test cases, each graded against an expected-behavior rubric by an LLM judge, from Corsac's LLM Observability & Evaluation Platform eval coverage.
How is the Prompt Hub And Prompt Management eval scored?+
The judge rubric: Grade against expected.ideal_behavior and expected.rubric. Per-criterion pass requires mean >= 4.0 and no criterion below 3.
How many test cases does this eval pack include?+
The Prompt Hub And Prompt Management pack for LangSmith LangSmith contains 9 test cases. 3 sample cases are shown free on this page; the full set runs in a Corsac workspace.
How do I run this eval?+
Sign up for Corsac, connect your model or agent endpoint, and run the Prompt Hub And Prompt Management pack as-is or after customizing thresholds. Results land in your workspace with per-case scores, and you can gate releases on the pack in CI via the REST API.
Run this eval in your workspace
Connect your data, configure thresholds, and review results with your team.