control-pane-ops

Runs cross-platform operations tasks across Slack, Stripe, Linear, GitHub, and databases with human approval gates.

20.2k|3.4k|Updated Oct 5, 2024
One-click install
npx skills add https://github.com/kortix-ai/suna --skill control-pane-ops
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: control-pane-ops
Source: https://github.com/kortix-ai/suna/tree/main/packages/starter/templates/marketplace/runtime/skills/control-pane-ops
Command: npx skills add https://github.com/kortix-ai/suna --skill control-pane-ops

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Operations teams juggle requests scattered across Slack threads that touch multiple platforms — billing lookups in Stripe, ticket filing in Linear, PR reviews in GitHub, and database queries — with no consistent access scoping or approval process for risky changes.

Core Features & Use Cases

  • Scoped multi-platform access: Each connected platform (database, Stripe, Linear, GitHub, sandbox) gets a defined read/write boundary, with read-only database credentials and brokered Stripe/Linear connectors.
  • Human approval gates: Money movement, production-data writes, merges, and account changes pause for explicit in-thread confirmation before execution.
  • Session isolation: Every Slack @-mention spawns a fresh session with no cross-thread state, and a scheduled heartbeat verifies each platform credential still authenticates.
  • Use Case: A teammate @-mentions the agent asking to refund a customer; the agent looks up the Stripe subscription, posts the exact refund action in the thread, and executes only after explicit approval.

Quick Start

@mention the agent in a Slack thread with an operations task such as looking up a customer's plan or reviewing a pull request.

Frequently Asked Questions about control-pane-ops

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run cross-platform operations from a Slack thread?

@-mention the agent in any Slack thread with your task. Each mention spawns a fresh isolated session that reads the full thread, scopes the request to the connected platforms, and replies with results in the same thread.

How do approval gates work for Stripe refunds and GitHub merges?

The agent posts the exact action — platform, specific change, and effect — in the thread and waits for an explicit reply before executing. It never proceeds on silence, and GitHub merges are never performed by the agent at all.

Can the agent write to the production database?

No. The database credential is a read-only role that cannot execute UPDATE, DELETE, or schema changes. The agent states the exact SQL needed in the thread and hands it to a human to run directly.

What does the heartbeat run check?

The scheduled heartbeat performs one lightweight read per platform — a Stripe account lookup, Linear viewer query, gh auth status, and a SELECT 1 — then posts a health alert to the ops channel naming any credential that failed.

Does session state carry over between Slack threads?

No. Every mention runs in its own fresh session and isolated sandbox with no memory of other threads or previous runs. The triggering thread is the only context and the only output channel.