Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Sign in
Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

A page, a video or an image online.

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

Error analysis for LLM applications v1.0.0

Use this resource to classify the failures of an LLM application from its traces before writing evals

New

  • 351 downloads
  • 117 likes
  • 203 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install llm-error-taxonomy-analysis

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent pulls real traces from Langfuse or a JSONL export, clusters them into named and counted failure modes, then has a human confirm the taxonomy. It then recommends binary pass/fail evals, only for the recurring named modes.

Mostly picked by

Updated November 2, 2026

Card for the “Error analysis for LLM applications” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: analysis phases, from trace collection to taxonomy
  • references/method.md: failure coding method
  • references/langfuse-traces.md: Langfuse trace extraction

Good fit for

  • A chatbot or AI agent misbehaves and you do not know what to measure
  • You have production traces to classify before writing evals

Not for

    Tags

    • error analysis
    • evals
    • traces

    About

    You say

    “Group the failures in these Langfuse traces and suggest the evals to write.”

    Card for the “Error analysis for LLM applications” skill

    Skill published by yonatangross under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to classify the failures of an LLM application from its traces before writing evals
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/judge-alignment.md3.2 KB
    • references/langfuse-traces.md6.7 KB
    • references/method.md5.3 KB
    • SKILL.md14.4 KB
    • test-cases.json2 KB

    Origin and licence

    Listed unmodified

    Author
    yonatangross
    Source
    github.com/yonatangross/orchestkit/tree/18a6aaa53d9971ef094b0758b08dfc606a1ea4bc/src/skills/error-analysis
    Licence
    MIT

    Notice

    « error-analysis » (yonatangross, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.