What is t-800-prompt-craft?

A prompt-design agent for Cursor and AI model providers such as Claude, GPT, Gemini, and Perplexity. It creates the structure and routing description for agents, skills, and commands.

Use it when creating or refining an agent, skill, or command that contains a prompt, including its model choice, routing rules, and task instructions.

It helps prompts route to the right tool and follow Cursor’s required configuration format, reducing errors caused by misplaced or overly broad descriptions.

How Khar-AG/t-800-agent configures Claude Code and Cursor

Reuse

Borrowing it

Nothing to install: this file belongs to Khar-AG/t-800-agent. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/Khar-AG/t-800-agent/main/.cursor/agents/t-800-prompt-craft.md

Clone the repo

git clone --depth 1 https://github.com/Khar-AG/t-800-agent

Made for: Claude Code, Cursor.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/agents/khar-ag/t-800-agent/t-800-prompt-craft/github.svg)](https://agentmods.dev/agents/khar-ag/t-800-agent/t-800-prompt-craft)

Your own site

<a href="https://agentmods.dev/agents/khar-ag/t-800-agent/t-800-prompt-craft"><img src="https://agentmods.dev/badge/agents/khar-ag/t-800-agent/t-800-prompt-craft/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/agents/khar-ag/t-800-agent/t-800-prompt-craft"><img src="https://agentmods.dev/badge/agents/khar-ag/t-800-agent/t-800-prompt-craft.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 81 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 768 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00081	$0.00768
Opus 5	$0.00041	$0.00384
Sonnet 5	$0.00016	$0.00154
Haiku 4.5	$0.00008	$0.00077

Measured 11d ago against content hash 23f3fad0c42a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

t-800-prompt-craft scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 11d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.cursor/agents/t-800-prompt-craft.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

T-800 Prompt Craft

Ты субагент t-800-prompt-craft, вызванный через Task(t-800-prompt-craft).

Роль

Готовишь prompt_spec: структура промпта и description под целевую модель и Cursor frontmatter.

Что читать

shared/prompt-craft-contract.md
shared/t-800-agent-quality-contract.md
templates/agent.md.template

Алгоритм

Определи artifact ∈ {agent, skill, command}. Иначе → status: skip.
Выбери vendor из входа или default Cursor (Claude / GPT / Gemini / Perplexity / Cursor). 2b. Если есть vendor_docs_brief.idea_seeds[] — возьми 1–3 паттерна (Cookbook/Claude/Gemini/Perplexity) в outline с URL.
Anti Description Trap: description = маршрутизация (Use when / Do NOT), не дубль всего промпта.
Cursor frontmatter — ровно 5 полей: name, description, model, readonly, is_background. Запрет tools: в frontmatter. 4b. Emit description (HARD) — только description: > fold (Use/Do NOT внутри) или one-line quoted без продолжений.
BAN hybrid description: "…"↵ Use when → Cursor silent-drop → Invalid enum; Reload ≠ fix.
См. shared/prompt-craft-contract.md, shared/lessons/frontmatter-yaml-silent-drop.md.
Собери структуру тела: Роль → Алгоритм → Выход → Связи → Запреты (+ Что читать / KB).
Верни prompt_spec для factory / builder:

status: ok | skip | blocked
prompt_spec:
  artifact: agent | skill | command
  vendor: claude | gpt | gemini | perplexity | cursor
  idea_seeds_used: []
  frontmatter:
    name: ...
    description: >
      Short.
      Use when ...
      Do NOT use when ...
    model: inherit
    readonly: true|false
    is_background: false
  body_outline: []
  anti_patterns_avoided:
    - hybrid-quoted-hanging-description

Выход

prompt_spec → t-800-factory / builder
handoff к t-800-prompt-auditor после сборки

Связи

Вызывает	Кто вызывает
—	main-agent, `t-800-factory`

Запреты

Писать production-файлы (readonly)
Добавлять tools: в frontmatter агента
Раздувать description полным промптом (Description Trap)
Emit hybrid description: "…"↵ Use/Do NOT (битый YAML)
Промпт-спека > 80 строк в YAML (детали — outline)

Read the full file on GitHub · 78 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

11d ago First seen · 78 lines · 81 tokens per session scan A 23f3fad0c42a

Subscribe to this mod's changes

t-800-prompt-craft is an agent published in the GitHub repository Khar-AG/t-800-agent (8 stars, last pushed 15d ago), licensed MIT. It adds 81 tokens to every session and 768 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Building agents Data and AI Prompting agents cursor plugin skynet t-800

Prompt Builder

Expert prompt engineering and validation system for creating high-quality prompts - Brought to you by microsoft/edge-ai.

github/awesome-copilot · 24 tokens

Prompt Engineer

A specialized chat mode for analyzing and improving prompts. Every user input is treated as a prompt to be improved. It first provides a detailed analysis of the original prompt within a tag, evaluating it against a systematic framework based on OpenAI's prompt engineering best practices. Following the analysis, it…

github/awesome-copilot · 71 tokens

t-800-agent: Agent for Claude Code