StarNet is a local-first desktop harness for creating AI agents and arranging them in a pixel-art space station whose layout represents their capabilities, permissions, and workflow. It lets people run real model calls and tools through multiple bounded agents, using their own provider keys or accounts and optionally messaging the station through chat services. The catalogue entries are skills that extend or support work performed by StarNet agents.
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 skills add androoAGI/starnet --skill calendar-schedulinggit clone --depth 1 https://github.com/androoAGI/starnetWrote 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/skills/androoagi/starnet/calendar-scheduling)<a href="https://agentmods.dev/skills/androoagi/starnet/calendar-scheduling"><img src="https://agentmods.dev/badge/skills/androoagi/starnet/calendar-scheduling/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.
<a href="https://agentmods.dev/skills/androoagi/starnet/calendar-scheduling"><img src="https://agentmods.dev/badge/skills/androoagi/starnet/calendar-scheduling.svg" alt="Reviewed on agentmods" width="80" 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 | $0.01018 |
| Opus 5.5 | $0.00019 | $0.00407 |
| Sonnet 5.5 | $0.00009 | $0.00204 |
| Haiku 4.5 | $0.00005 | $0.00102 |
Grade A, and why
calendar-scheduling 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 12d 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 — 41 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Scheduling goes wrong in two ways: a time that was not actually free, and a time zone nobody said out loud. Offer only slots you checked, name the zone on every time, and change nothing on the calendar until the Commander says so.
Method
- Load the Commander's rules.
notebook.readfor scheduling preferences: home time zone, working hours, days off, the gap wanted between meetings, no-meeting blocks, default length. Ask once for a missing one this request depends on, then save it withnotebook.writeso you never ask again. - Pin the request. Who, how long, by when, in person or a call, and the time zone of every other person. If someone's zone is unknown, ask; never assume it is the Commander's.
- Find the calendar tools. Use the connected Google account's calendar tools (find them with
tool.search); never guess a tool name. If no calendar is connected, say so, ask the Commander to connect Google in ABILITIES, and until then work only from times they give you, labelled "not checked against the calendar". - Read availability. Pull every event in the window, across each calendar the Commander uses. Tentative events count as busy unless told otherwise. All-day events marked busy (travel, leave) block the day; all-day markers such as birthdays do not. For in-person meetings add travel time on both sides.
- Build the candidates. Free time, inside working hours, with the gap kept before and after, and inside reasonable hours for the other people (08:00 to 18:00 their time unless told otherwise). Convert between zones with
code.run, not in your head, and check for a daylight-saving change inside the window. - Rank three. Best first: inside everyone's working hours, gaps kept, no focus block split, soonest. Give each a one-line reason. Fewer than three good ones: offer what exists and say which rule would have to bend for more.
- Write every time with its zone. Weekday, date, start and end, zone name and UTC offset, repeated for each party: "Tue 14 Oct, 15:00-15:30 London (BST, UTC+1) = 10:00-10:30 New York (EDT, UTC-4)".
- Draft the invite and hold it. Title, time with zone, length, attendees, place or call link, a two-line agenda. Show it. Say plainly that creating it will email the guests.
- On the go-ahead, act once and read back. Create the event with the calendar tool, then read the event back and confirm the stored start, zone and attendees match the draft. Report a mismatch instead of "booked".
- Handle conflicts in the open. For a double-booking, or a request that only fits by moving something: show both events, say which looks easier to move and why (fewer people, internal, repeats weekly), propose the change, and wait.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 12d ago First seen · 41 lines · 47 tokens per session scan A 0ed58792e5d6
calendar-scheduling is a skill published in the GitHub repository androoAGI/starnet (1,468 stars, last pushed 3d ago), licensed MIT. It adds 47 tokens to every session and 1,018 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-10-04.
Other skills, from other repositories
orbit-general
Open Orbit briefing skill — selected by the Orbit pipeline when the user has two or more connectors connected. Pulls the past 24 hours of activity from every authenticated connector (GitHub, Linear, Notion, Slack, 飞书, Calendar, Gmail, Drive, Sentry, Vercel, …) and renders a single adaptive bento-grid dashboard at the…
orbit-gmail
Open Orbit briefing skill — selected by the Orbit pipeline when Gmail is the user's only connected connector, or when the user explicitly scopes their daily digest to Gmail. Pulls the past 24 hours of inbox activity (replies awaited, mentions, cc, auto- categorized bulk) from the user's authenticated Gmail connection…
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
discovery-question-form
Structured clarification form for unresolved material requirements.
cron
Use this skill only for scheduled or recurring tasks. Manage jobs with qwenpaw cron list/create/get/state/update/pause/resume/delete/run, and always pass --agent-id explicitly.
cron
A procedure for creating and managing tasks that run automatically at a future time or repeat on a schedule. It uses QwenPaw's scheduling commands and requires an agent identifier.