Skip to content

CODE-SAURABH/OpenSkills / ai-engineer

ai-engineer

Unverified

Expert AI/LLM engineering guidance. Use when the user asks about RAG, OpenAI, Azure OpenAI, prompt engineering, AI agents, MCP, vector databases, or evaluation pipelines.

FileKindSize
SKILL.mdskill md34.4 KB
references/evaluation-design.mdmarkdown901 B

references/evaluation-design.md

View on GitHub
# Evaluation Design

## Build an Eval Set

Use representative, consented examples. Include ordinary requests, boundary cases, known failures, adversarial inputs, and cases where the correct response is to abstain or ask for clarification. Keep the production-like input, expected behavior, and any safety constraints together.

## Select Metrics

Choose metrics that correspond to the user outcome: task success, groundedness, structured-output validity, latency, cost, or safe refusal. Record the scorer, rubric, and pass condition. Avoid a single aggregate score when a safety or reliability metric needs its own gate.

## Release Gate

Compare a candidate against a fixed baseline on the same eval set. Define acceptable regressions before running it, inspect failures by category, and retain examples that reveal real defects. Treat human review as required for subjective or high-impact outputs.
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 197kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall