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/perniemann/pncore/pn-communication-contractgit clone --depth 1 https://github.com/perniemann/pnCoreWrote 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/perniemann/pncore/pn-communication-contract)<a href="https://agentmods.dev/rules/perniemann/pncore/pn-communication-contract"><img src="https://agentmods.dev/badge/rules/perniemann/pncore/pn-communication-contract.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 | $0.00047 | $0.00316 |
| Opus 5 | $0.00023 | $0.00158 |
| Sonnet 5 | $0.00009 | $0.00063 |
| Haiku 4.5 | $0.00005 | $0.00032 |
Grade A, and why
pn-communication-contract 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 5d 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.
What it actually says
Communication contract
Agent-requested (not always-on). Load for verbose or sloppy chat, compression asks, or response aliases. Examples: pn-core://reference/communication-contract.md. Aliases: get_skill("pn-response-aliases").
Positive patterns
- Put the answer first; details after.
- State each fact once. Match detail to the ask.
- Use plain, specific language. Prefer one short paragraph over two.
Negative patterns
- No hedging: "possibly", "should", "could", "might" as soft claims.
- Avoid filler: "you're absolutely right", "great question", "worth stating plainly", "load-bearing".
- No decorative headings, emoji cheerleading, or unsolicited praise.
- Do not restate the user's question before answering.
Boundaries (point, do not duplicate)
- Deliver only the requested scope — no surprise cleanup or adjacent features.
- Completion claims need evidence (
pn-verification-before-completion). - Commit trailers:
pn-no-cursor-commit-trailers. Build/phase gates:pn-build-gate.
Aliases
Whole-token only: scr, eli, foc, ref, scp — expand via pn-response-aliases.
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.
- 5d ago First seen · 32 lines · 47 tokens per session scan A bda8db89b8ea
pn-communication-contract is a cursor rule published in the GitHub repository perniemann/pnCore (0 stars, last pushed 2d ago), licensed MIT. It adds 47 tokens to every session and 316 once invoked, about $0.0002 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.
Other cursor rules, from other repositories
t-800-mandatory-routing
ОБЯЗАТЕЛЬНО: создание и правка субагентов, skills, commands, rules, hooks — только через T-800 Agent (цех Skynet). Сначала intake? + scout + research DEEP + brain-lead, затем t-800-factory. Не собирать артефакты ad-hoc в main chat.
t-800-operator-routing
Маршрутизация к T-800 Operator для русскоязычных новичков: первый запуск Cursor, вопросы про Agent/Ask/Plan/Debug, rules, skills, MCP, автоматизацию, просьбы объяснить «на пальцах». Делегируй Task(t-800-operator), не дублируй длинные объяснения в основном чате.
t-800-factory-routing
Маршрутизация к T-800 Factory: создание субагентов, проектирование плагина, интеграция агентов, реестр, граф связей, масштабирование 100+ agents. Делегируй Task(t-800-factory), не создавай агентов ad-hoc в main chat.
t-800-knowledge-refresh
Обновление базы знаний T-800: когда появились новые функции Cursor, устарел manifest (>30 дней), есть UPDATE-QUEUE.md с pending, пользователь спрашивает про фичу которой нет в KB, или после changelog Cursor. В проекте Cursor Junior делегируй обслуживание Task(t-800-maintainer).
t-800-cloud-hub-routing
Маршрутизация Cloud Hub Automation Setup: при /t800-cloud-hub, /t800-hub-setup или правках /cloud-hub/ → Task(t-800-cloud-hub-lead). Не always-on.
ponytail
Ponytail, lazy senior dev mode. Always pick the simplest solution that works.