Building LLM Backed Features

Builds application features on top of an LLM API in Python — prompt surfaces that drift apart, structured outputs, context budgeting, model/config wiring, filtered evaluator sets, keeping live-model calls out of default CI, evaluating stochastic output, and presenting model output truthfully. Use when adding an LLM call to an app, designing structured outputs or prompt templates, filtering audit rules, wiring model config, or testing/evaluating LLM-backed analysis.

115 stars 18 forksAI, LLMs & Agents Plugin marketplace

Install

Claude Code plugin
/plugin marketplace add wdm0006/python-skills

The repo is a Claude Code plugin marketplace. Run /plugin afterwards to pick the plugin that includes this skill.

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add wdm0006/python-skills --skill building-llm-backed-features
Manual install
git clone --depth 1 https://github.com/wdm0006/python-skills.git /tmp/python-skills
mkdir -p ~/.claude/skills
cp -r /tmp/python-skills/skills/python/llm-features ~/.claude/skills/building-llm-backed-features

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Building LLM-Backed Features

Calling a model is easy; shipping a feature around one is where the bugs live.
The model is a non-deterministic dependency whose *inputs are spread across your
codebase* and whose *outputs get presented to users as facts*. This skill covers
the failure modes that recur in that gap. (Building a server that exposes tools
**to** a model is the other direction — see
**[../mcp-servers/SKILL.md](../mcp-servers/SKILL.md)**.)

## Every string the model reads is prompt surface

Wording that shapes model behavior does not live only in the prompt template. It
typically lives in four places that drift apart independently:

| Layer | Looks like | Actually is |
|-------|-----------|-------------|
| System/user prompt template | prompt | prompt |
| Structured-output field descriptions | docstrings | **prompt** — the model reads them |
| Strings you interpolate into rendered results | formatting | user-facing claims |
| Template/API disclaimers | copy | the user's only caveat |

The second row is the trap. A `Field(description=...)` is part of the schema sent
to the model, so softening the prompt while leaving a categorical field
description in place changes nothing about the output.
Read the full skill on GitHub

Details

Skill name
building-llm-backed-features
Path
skills/python/llm-features/SKILL.md
License
MIT
Repo updated
18 November 2026

FAQ

What does the Building LLM Backed Features skill do?

Builds application features on top of an LLM API in Python — prompt surfaces that drift apart, structured outputs, context budgeting, model/config wiring, filtered evaluator sets, keeping live-model calls out of default CI, evaluating stochastic output, and presenting model output truthfully. Use when adding an LLM call to an app, designing structured outputs or prompt templates, filtering audit rules, wiring model config, or testing/evaluating LLM-backed analysis.

How do I install the Building LLM Backed Features skill?

Claude Code plugin: /plugin marketplace add wdm0006/python-skills. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Building LLM Backed Features popular?

It lives in wdm0006/python-skills, which has 115 GitHub stars and 18 forks.

Related skills

Auditing Python Security

115

wdm0006/python-skills

Audits Python libraries for security vulnerabilities using Bandit, pip-audit, Semgrep, and detect-secrets. Identifies SQL injection, command injection, hardcoded credentials, secrets exposed through tracebacks, weak cryptography, and insecure deserialization. Use when reviewing library security, setting up security scanning in CI, or implementing secure coding patterns.

Security18

Building Accessible Interfaces

115

wdm0006/python-skills

Makes web and app UIs usable by keyboard, screen-reader, low-vision and motion-sensitive users — semantic HTML before ARIA, labelled controls, visible focus, WCAG AA contrast measured rather than eyeballed, touch targets, reduced motion, and automated axe checks plus a manual keyboard pass in CI or before every UI PR. Use when building or restyling a screen, adding a form, modal, menu or custom control, choosing colors, or reviewing a UI change.

Frontend & Design18

Building Browser Extensions

115

wdm0006/python-skills

Builds and tests Manifest V3 browser extensions — service-worker message channels that survive teardown, chrome.storage read-modify-write safety, settings validation in the popup, and a Jest chrome mock that can actually fail. Use when writing or reviewing a background service worker, a popup that reads/writes chrome.storage, an extension's Jest suite, or debugging an extension whose UI silently stops responding.

Software Development18

Building Python MCP Servers

115

wdm0006/python-skills

Builds robust Python MCP (Model Context Protocol) servers with FastMCP — tool design, error contracts, event-loop-safe blocking work, subprocess/CLI wrapping, single-file vs packaged distribution, global-state-free testing, and prompt-injection awareness. Use when writing an MCP server, exposing a tool or CLI to an LLM client, debugging tool registration/packaging, or testing MCP tools.

AI, LLMs & Agents18

Xhs AI Tools Writer

45

mini20201314-crypto/xhs-ai-tools-writer

小红书 AI 工具向编稿工作流。用于写/改小红书笔记:ChatGPT、Claude、Cursor、 Midjourney、效率插件等真实测评与场景干货。触发词:小红书、xhs、编稿、 笔记、标题、AI工具测评、Cursor、ChatGPT种草、科技干货。 Use when writing Xiaohongshu notes about AI tools, productivity apps, or tech how-tos.

AI, LLMs & Agents20

H3

46

heavy3-ai/code-audit

Heavy3 Code Audit - Multi-model code review for coding agents (Sponsored by Heavy3.ai)

AI, LLMs & Agents15

Browse the most popular Claude skills, all ai, llms & agents skills, or write your own with the skills generator.