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 sonichi/sutando --skill connect-appsgit clone --depth 1 https://github.com/sonichi/sutandoWrote 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/sonichi/sutando/connect-apps)<a href="https://agentmods.dev/skills/sonichi/sutando/connect-apps"><img src="https://agentmods.dev/badge/skills/sonichi/sutando/connect-apps/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/sonichi/sutando/connect-apps"><img src="https://agentmods.dev/badge/skills/sonichi/sutando/connect-apps.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 | $0.00212 | $0.08945 |
| Opus 5.5 | $0.00085 | $0.03578 |
| Sonnet 5.5 | $0.00042 | $0.01789 |
| Haiku 4.5 | $0.00021 | $0.00894 |
Grade A, and why
connect-apps 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 6d 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 — 490 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Connect apps
The owner's apps (Google Calendar, Gmail, Google Meet, Google Drive, Slack, Linear, Notion, GitHub,
and a thousand more) are reached through the sutando-station MCP server. A connection made
mid-session works at once; nothing here ever needs a restart.
When an app isn't connected, the flow looks like this in the owner's chat:
- You: "Your Google Calendar isn't connected yet, so I can't peek at your schedule. Want to hook it up?"
- A card: the app's icon and name with a Connect button.
- You: "Once that's done I'll be able to check what's coming up for you." The task ends here.
- After sign-in you continue by yourself: "Your Google Calendar is connected and ready to go. Here's your Friday: ..."
In a room with other people, lines 1 to 3 are not messages: they are a private card under the owner's message, inside their "Only visible to you" activity card, which no other member receives. The room sees nothing about connecting, only your reply once the request is done (step 3b).
Every request gets exactly one answer: the steps below check for a wait before answering.
Calendar, contacts and reminders: the connector first, the native Mac apps never on your own.
Order: (1) the Station connector (composio_find {"apps": ["google calendar"]} → composio_exec,
toolkit googlecalendar); (2) if it is not connected, the owner's own tools when they are in your
tool list (mcp__claude_ai_Google_Calendar__*); (3) otherwise the Connect card below, or ask the
owner. Never fall back to the macOS Calendar, Reminders or Contacts app by yourself — it raises a
permission prompt on the owner's screen (the native-pim-guard hook denies the command). Only when
the owner asked for the local app: skills/macos-tools scripts with --owner-asked. Once the owner
denied that permission, never re-prompt.
Tools
mcp__sutando-station__composio_find{query?, apps?: [names], toolkit?, limit?}returns JSON{"apps":[{toolkit, name, icon_url, connected, auth_mode, coming_soon}], "actions":[{toolkit, action, description, inputSchema}], "next_step"?}.appsmatches app names ("google calendar", "linear"); actions are listed for connected apps and for an exacttoolkitslug.mcp__sutando-station__composio_exec{toolkit, action, arguments}runs one action. An error whose message containsconnect_requiredmeans the app is not connected.mcp__sutando-station__station_find/station_callreach the owner's cloud tools, including one activated a minute ago. See "Paid cloud tools" below.- The helper script, next to this file:
What ships with it
4 files 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.
- 6d ago Changed · +8 lines b984c02c2896
- 7d ago Changed · +10 lines acb6c2e7ac11
- 19d ago Changed · +21 lines 9cc302327002
- 20d ago Changed · +42 lines 0de3fffe94a2
- 24d ago Changed · +13 lines 0740e19c4190
- 29d ago Changed · +155 lines · +68 tokens per session a72605aaccb3
- 1mo ago First seen · 241 lines · 144 tokens per session scan A 64a96bb9db8c
connect-apps is a skill published in the GitHub repository sonichi/sutando (396 stars, last pushed today), licensed MIT. It adds 212 tokens to every session and 8,945 once invoked, about $0.0008 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 2026-09-16.
Other skills, from other repositories
gmail
Manage Gmail email — drafting, sending, organizing, filters, vacation replies, and inbox analysis.
schedule
Reminders, automations, and recurring page, dashboard, or status monitoring (cron, RRULE, or fire-at).
inbox-management
Ongoing Gmail inbox management via scheduled runs. Archives known noise, flags urgent items, drafts replies in-thread (never auto-sends), and catches stale follow-ups. Empty polls spend no model tokens. Starts in flag-only mode.
inbox-cleanup
Run a high-recall, multi-pass email inbox cleanup. Pattern-based subject queries catch 25x more archivable email than sender scans alone. Includes urgency triage, classification signals, and post-cleanup filter setup.
notifications
Send notifications through the unified notification router.
outlook
Manage Outlook email — drafting, sending, organizing, rules, vacation replies, and inbox analysis.