# Super Calendar LLM Instructions

This file is for automated agents, LLMs, crawlers, coding assistants, and integration tools.

It is the short discovery map. It is not the full endpoint schema and not the full autonomous-agent contract.

Prefer machine-readable contracts over guessing.

## Product identity

Super Calendar converts time-related data into portable calendar-ready events.

Primary abstraction:

```txt
source data -> normalized event semantics -> portable output surface
```

Super Calendar is not only a calendar website. It is a temporal-data transformation platform for humans, applications, devices, and agents.

## More complete instructions

For the full LLM-readable instruction document, use:

```txt
/agents/llms-full.txt
```

For autonomous-agent behavior, use:

```txt
/agents/contract.md
```

For agent resource discovery, use:

```txt
/agents/index.json
```

For capability planning, use:

```txt
/agents/capabilities.json
```

For structured landing/product interpretation, use:

```txt
/agents/landing.index.json
```

For exact HTTP routes, parameters, response schemas, and supported formats, use:

```txt
/api/swagger/json
```

For low-bandwidth I/O node addressing, use:

```txt
/io/ioapi.json
```

For runtime language availability and file mapping, use:

```txt
/common/i18n/manifest.json
```

For human calendar-client import instructions, use:

```txt
/importing/
```

## Main domains

Known public or planned domains may include:

```txt
weather
moon
holidays
exchange
crypto
earnings
seasons
religion
events
trivia
custom
spaces
```

Do not assume this list is complete forever.

Do not assume every domain supports the same parameters, formats, locales, authentication state, preview behavior, or I/O fields.

## Core agent rules

- Read `/agents/llms-full.txt` for full LLM instructions.
- Read `/agents/contract.md` before autonomous execution.
- Read `/agents/capabilities.json` before choosing a domain or output surface.
- Read `/api/swagger/json` before constructing HTTP URLs.
- Read `/io/ioapi.json` before constructing I/O node URLs.
- Read `/common/i18n/manifest.json` before claiming a language is available.
- Do not invent routes, parameters, formats, domains, lifecycle states, billing, authentication, storage, or personalization.
- Do not infer language, country, timezone, location, or identity from IP address.
- Treat private WebCal, ICS, widget, and feed URLs as access credentials.
- Do not treat crawler-visible loading placeholders as final product state when pages intentionally hydrate client-side.

## Output surfaces

Depending on the domain and contract, output surfaces may include:

```txt
webcal
ics
json
xml
text
ansi
html
csv
tsv
ndjson
jsonl
io
mcp
gateway
```

Do not claim a surface works for a domain until OpenAPI, I/O, MCP, or the capability registry confirms it.

## Runtime language note

HTML may initially render in English and hydrate to another language client-side. Initial English HTML is not proof that other languages are unsupported.

Use `/common/i18n/manifest.json` for runtime language availability.

## Privacy note

For anonymous visitors: do not infer country, language, geolocation, timezone, or identity from IP; do not persist anonymous browser language; do not fingerprint visitors; explicit user choice always wins.