Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add rules/khar-ag/t-800-agent/t-800-knowledge-refreshgit clone --depth 1 https://github.com/Khar-AG/t-800-agentWrote 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.
[](https://agentmods.dev/rules/khar-ag/t-800-agent/t-800-knowledge-refresh)<a href="https://agentmods.dev/rules/khar-ag/t-800-agent/t-800-knowledge-refresh"><img src="https://agentmods.dev/badge/rules/khar-ag/t-800-agent/t-800-knowledge-refresh.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00000 | $0.00717 |
| Opus 5 | $0.00000 | $0.00358 |
| Sonnet 5 | $0.00000 | $0.00143 |
| Haiku 4.5 | $0.00000 | $0.00072 |
Grade A, and why
t-800-knowledge-refresh 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
T-800 — обновление базы знаний
Контракт: shared/knowledge-update-contract.md
Когда обновлять (триггеры)
- Планово:
manifest.json→last_full_syncстарше 30 дней - Событийно: новый релиз / changelog Cursor, новая страница в docs
- По запросу: пользователь спрашивает про функцию, которой нет в
knowledge-base/INDEX.md - После sync: в
UPDATE-QUEUE.mdесть статусыnewилиchanged
Алгоритм (в проекте Cursor Junior)
- Делегировать работу через
Task(t-800-maintainer), если нужно не просто ответить пользователю, а обновить KB - Прочитать
knowledge-base/manifest.jsonиUPDATE-QUEUE.md - Ручной sync: fetch URL из
manifest.json→raw/, статусы вUPDATE-QUEUE.md - Запустить:
.\scripts\audit-coverage.ps1 - Открыть
UPDATE-QUEUE.md— обработать pending - Для каждого изменения:
- Упростить текст для новичка (не копировать docs 1:1)
- Добавить: Простыми словами, Пошагово, Схема, Частые ошибки
- Обновить
last_syncedв frontmatter карточки
- Записать в
knowledge-base/CHANGELOG.md - Обновить
INDEX.md/glossarium.md/profiles//wizards/при новых сценариях .\scripts\install-plugin.ps1— синхронизировать глобальную установку.\scripts\verify-install.ps1— проверить, что subagent не превратился в skill
Для T-800 Operator (субагент, readonly)
T-800 Operator не запускает sync сам в readonly-режиме. Он:
- Проверяет дату
last_syncedкарточки иmanifest.json - Если данные старше 30 дней — предупреждает: «информация может быть устаревшей»
- Если темы нет в KB — даёт официальную ссылку + совет обновить базу
- Предлагает владельцу проекта:
/t-800-sync(ручной sync KB)
Запреты
- Не перезаписывать упрощённые карточки сырым HTML из
raw/без ревью - Не удалять ручные аналогии и mermaid при merge
- Не коммитить секреты из docs
Команды
.\scripts\audit-coverage.ps1— проверить покрытие manifest карточками.\scripts\install-plugin.ps1— обновить ~/.cursor/.\scripts\verify-install.ps1— проверить установку
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.
- 2d ago First seen · 58 lines · 0 tokens per session scan A 07c39204c862
t-800-knowledge-refresh is a cursor rule published in the GitHub repository Khar-AG/t-800-agent (8 stars, last pushed 9d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 717 tokens. 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 2026-09-03.
Other cursor rules, from other repositories
pn-aesthetics-baseline
Non-generic UI baseline for user-facing surfaces; aligns with .pncore-design.md and pn-core aesthetics reference when MCP is available.
pn-build-gate
For build/design/scaffold requests, load pn-build or pn-design; discovery and skeptic mandatory. For fix/tweak, use skills directly.
pn-godot
Godot 4.x coding conventions: GDScript typing, signal patterns, scene composition, shader style, naming, Resource patterns, multiplayer RPC safety, GDExtension registration. Use when editing .gd, .tscn, .tres, .gdshader, or project.godot files.
pn-mcp-proactive
When MCP is available, use pn-core tools proactively when tasks match pn skills or agents; use Octocode for code research when available.
pn-nextjs
Next.js best practices. Data loading, server/client boundaries, streaming, mutations, and performance. For Next app/ or pages/, pn-react also applies (core React patterns); content is complementary.
pn-php-backend
PHP backend style, env, and error handling. Use for API and server code (Laravel, Slim).