Prompt Optimizer Toolkit

LLM Prompt Engineering Toolkit - 提示词分析、优化、模板化与质量评分 | Analyze, optimize, template, and score your LLM prompts with built-in engineering strategies

Security Scan

The skill's files, runtime instructions, and requested resources are consistent with a local prompt-analysis/optimization toolkit that only needs Python and writes local prompt storage.

Purpose & Capability

Name/description match the included Python code and examples. The code implements analysis, optimization, templating, scoring, and a local prompt library — all coherent with the stated purpose.

Instruction Scope

Runtime instructions are limited to running the provided Python module and installing Python deps. The code performs local file I/O (default storage prompts.json or user-specified path) to persist prompts; users should expect prompt data to be written to the working directory. There are no network calls, secret reads, or broad filesystem scans in the SKILL.md or code.

Install Mechanism

This is effectively instruction-only aside from a python dependency step referencing requirements.txt (which only pins pytest). No external downloads, tarballs, or obscure installers are used.

Credentials

The skill requires only python3 (declared) and no environment variables or credentials. The code does not reference hidden env vars or external service keys.

Persistence & Privilege

The skill does not request always:true and does not modify other skills or system-wide configuration. It stores prompts in a local JSON file under the chosen path (default prompts.json).

Assessment

This skill appears coherent and self-contained. Before installing: (1) be aware it writes a JSON file (default prompts.json) in the working directory — avoid using it in folders with sensitive data or set a custom storage_path; (2) it has no network calls or secret handling, and requirements.txt only includes pytest; (3) tests add an absolute workspace path to sys.path (expected for the test harness) — not a runtime concern; (4) run it in a sandbox or review the prompt storage if you plan to store sensitive prompts. Overall it is consistent with its description.

功能

• Analyze — 多维度分析提示词质量（清晰度、具体性、上下文、约束、示例、角色）
• Optimize — 自动检测最佳优化策略并应用
• Template — 内置5种经典提示词模板（CoT、Few-shot、Role-play、Structured、Constraints）
• Library — 本地提示词库管理，支持标签检索
• Score — 1-10分综合质量评分

使用

from scripts.prompt_optimizer import PromptOptimizer, PromptLibrary

optimizer = PromptOptimizer()

# 分析提示词
analysis = optimizer.analyze("Write a story about a robot")
# → {'clarity_score': 7, 'specificity_score': 5, 'overall_score': 4, ...}

# 自动优化
better = optimizer.optimize("Write a story about a robot")
# → 添加角色、结构、约束

# 使用模板
prompt = optimizer.apply_template("role_playing", domain="creative writing", task="Write a sci-fi story")

# 提示词库
lib = PromptLibrary()
lib.save("coding-helper", "You are a senior Python developer...", tags=["coding", "python"])

CLI

python3 scripts/prompt_optimizer.py