Official Verified

llm

Build and evaluate LLM prompts. Use when crafting system prompts, comparing variants, estimating tokens, or managing prompt templates.

skill-install — Terminal

Install via CLI (Recommended)

clawhub install openclaw/skills/skills/bytesagain/llm

Download Source Code (.zip)

llm

LLM Prompt Engineering Toolkit. Build structured prompts from role/context/task components, compare prompt variations side by side, estimate token counts, manage reusable prompt templates, chain multi-step prompts, and evaluate prompt quality with a scored breakdown. All commands run locally in bash with no API keys or network access required.

Commands

`prompt` — Build a Structured Prompt

Assembles a prompt from modular components: role, context, task, constraints, and output format. The --task flag is required; all others are optional.

Flags:

--role <text> — Define the AI's persona (e.g., "senior developer")
--context <text> — Provide background information
--task <text> — (required) The main instruction
--constraints <text> — Rules or limitations
--format <text> — Desired output format

bash scripts/script.sh prompt --role "senior developer" --context "Python Flask app" --task "write unit tests"
bash scripts/script.sh prompt --task "summarize this article" --constraints "max 3 sentences" --json

`compare` — Compare Prompt Variations

Compare two or more prompt files side by side. Shows each variant with word/line/char/token stats, then a diff --side-by-side of the first two variants, plus a summary table.

Flags:

--prompts <file1> <file2> [file3...] — Two or more prompt text files to compare

bash scripts/script.sh compare --prompts prompt_a.txt prompt_b.txt
bash scripts/script.sh compare --prompts v1.txt v2.txt v3.txt

`tokenize` — Estimate Token Count

Estimate the token count for a given text using a cl100k_base-compatible heuristic. Reports characters, words, lines, and estimated tokens.

Input methods:

--input <text> — Inline text string
--file <path> — Read from a file
Pipe via stdin

bash scripts/script.sh tokenize --input "Your prompt text here"
bash scripts/script.sh tokenize --file prompt.txt
echo "some text" | bash scripts/script.sh tokenize
bash scripts/script.sh tokenize --file prompt.txt --json

`template` — Manage Prompt Templates

Save, list, load, and delete reusable prompt templates. Templates are stored as .txt files in ~/.llm-skill/templates/.

Actions:

--save <name> --file <path> — Save a template from a file (or pipe via stdin)
--list — List all saved templates with sizes
--load <name> — Output the contents of a saved template
--delete <name> — Remove a saved template

bash scripts/script.sh template --save my_template --file prompt.txt
bash scripts/script.sh template --list
bash scripts/script.sh template --list --json
bash scripts/script.sh template --load my_template
bash scripts/script.sh template --delete my_template
echo "Write a haiku about {{topic}}" | bash scripts/script.sh template --save haiku

`chain` — Multi-Step Prompt Chains

Read Full Documentation on GitHub

Metadata

Author@bytesagain

Stars3500

Updated2026-03-27

View Author Profile

AI Skill Finder

Not sure this is the right skill?

Describe what you want to build — we'll match you to the best skill from 16,000+ options.

Find the right skill

Add to Configuration

Paste this into your clawhub.json to enable this plugin.

{
  "plugins": {
    "official-bytesagain-llm": {
      "enabled": true,
      "auto_update": true
    }
  }
}

Related Skills

sealvera

Tamper-evident audit trail for AI agent decisions. Use when logging LLM decisions, setting up AI compliance, auditing agents for EU AI Act, HIPAA, GDPR or SOC 2, or when a user asks about AI decision audit trails, explainability, or SealVera.

ahessami123 4473

rocm_vllm_deployment

Production-ready vLLM deployment on AMD ROCm GPUs. Combines environment auto-check, model parameter detection, Docker Compose deployment, health verification, and functional testing with comprehensive logging and security best practices.

alexhegit 4473

project-evaluator

描述一个项目想法，AI 从市场/技术/商业/风险四个维度系统评估，输出评估报告、竞品速查、MVP建议，帮你决策「值不值得做」。

antonia-sz 4473

opentangl

Not a code generator — an entire dev team. You write the vision, it ships the code. Autonomous builds, PRs, reviews, and merges across multiple repos. Point it at any JS/TS project and a product vision. It plans features, writes code, verifies builds, creates PRs, reviews diffs, and merges — autonomously. Manages multiple repos as one product. Use when you want to ship code without writing it. AI code generation, autonomous development, workflow automation, multi-repo orchestration, TypeScript, JavaScript, GitHub, OpenAI, Anthropic, Claude, GPT, LLM, devtools, CI/CD, pull requests, code review.

8co 4473

interview-evaluation-report

面试评估报告。触发场景：用户提供面试记录或面试笔记，要求生成结构化评估报告。

51mee-com 4473

llm