菜单
Build a custom browser-based annotation interface tailored to your data for reviewing LLM traces and collecting structured feedback. Use when you need to build an annotation tool, review traces, or collect human labels.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Calibrate an LLM judge against human labels using data splits, TPR/TNR, and bias correction. Use after writing a judge prompt (write-judge-prompt) when you need to verify alignment before trusting its outputs. Do NOT use for code-based evaluators (those are deterministic; test with standard unit tests).
Help the user systematically identify and categorize failure modes in an LLM pipeline by reading traces. Use when starting a new eval project, after significant pipeline changes (new features, model switches, prompt rewrites), when production metrics drop, or after incidents.
Audit an LLM eval pipeline and surface problems: missing error analysis, unvalidated judges, vanity metrics, etc. Use when inheriting an eval system, when unsure whether evals are trustworthy, or as a starting point when no eval infrastructure exists. Do NOT use when the goal is to build a new evaluator from scratch (use error-analysis, write-judge-prompt, or validate-evaluator instead).
Guides evaluation of RAG pipeline retrieval and generation quality. Use when evaluating a retrieval-augmented generation system, measuring retrieval quality, assessing generation faithfulness or relevance, generating synthetic QA pairs for retrieval testing, or optimizing chunking strategies.
Create diverse synthetic test inputs for LLM pipeline evaluation using dimension-based tuple generation. Use when bootstrapping an eval dataset, when real user data is sparse, or when stress-testing specific failure hypotheses. Do NOT use when you already have 100+ representative real traces (use stratified sampling instead), or when the task is collecting production logs.
Design LLM-as-Judge evaluators for subjective criteria that code-based checks cannot handle. Use when a failure mode requires interpretation (tone, faithfulness, relevance, completeness). Do NOT use when the failure mode can be checked with code (regex, schema validation, execution tests). Do NOT use when you need to validate or calibrate the judge — use validate-evaluator instead.
| name | build-review-interface |
| description | Build a custom browser-based annotation interface tailored to your data for reviewing LLM traces and collecting structured feedback. Use when you need to build an annotation tool, review traces, or collect human labels. |
Build an HTML page that loads traces from a data source (JSON/CSV file), displays one trace at a time with Pass/Fail buttons, a free-text notes field, and Next/Previous navigation. Save labels to a local file (CSV/SQLite/JSON). Then customize to the domain using the guidelines below.
Format all data in the most human-readable representation for the domain. Emails should look like emails. Code should have syntax highlighting. Markdown should be rendered. Tables should be tables. JSON should be pretty-printed and collapsible.
<details> toggle.Annotate at the trace level. The reviewer judges the whole trace, not individual spans.
Once you have established failure categories from error analysis, you can later add predefined failure mode tags as clickable checkboxes, dropdowns or picklists so reviewers can select from known categories in addition to writing notes. But don't add these in the initial build.
Arrow keys = Navigate traces
1 = Pass 2 = Fail
D = Defer U = Undo last action
Cmd+S = Save Cmd+Enter = Save and next
Build the app to accept traces from any source (JSON/CSV file). Keep sampling logic outside the app in a separate script. Start with random sampling.
Reference panel: Toggle-able panel showing ground truth, expected answers, or rubric definitions alongside the trace.
Filtering: Filter traces by metadata dimensions relevant to the product (channel, user type, pipeline version).
Clustering: Group traces by metadata or semantic similarity. Show representative traces per cluster with drill-down.
After building the interface, verify it with Playwright.
Visual review: Take screenshots of the interface with representative trace data loaded. Review each screenshot for:
Functional test: Write a Playwright script that performs a full annotation workflow: