Skip to content
Skills Library
AI & agents · Automation

agent-ops-command

/agent-ops-command

Run the daily operating rhythm with agents that sweep every system each morning, surface what needs a decision, and keep nothing waiting silently. Use when running a business day to day and things fall through the cracks between desks.

Repository
AI-SKILLS ↗
Licence
MIT
Last repository change
05.09.2026

Assessment and security

90/ 100Class: good
Format25 / 25
Frontmatter with name and description, valid name, description up to 1,024 characters, name matching the folder.
Discoverability25 / 25
Description explains when to use the skill, has a reasonable length and uses third person.
Content25 / 35
At least 150 words and at most 500 lines, headings, step lists, examples or code.
Hygiene15 / 15
No placeholders such as TODO, no Windows paths, with a licence.
Riskno signals detected in the scan
Files in this package
1
Scripts
0
HTML comments
0
URLs in SKILL.md
0
Deprecation

No signals detected. Repository changed on 05.09.2026, without references to retired models or APIs.

Licence

MIT, open: use and modify while retaining author attribution.

What it does and when to use it

Run the daily operating rhythm with agents that sweep every system each morning, surface what needs a decision, and keep nothing waiting silently. Use when running a business day to day and things fall through the cracks between desks.

Original author description

Run the daily operating rhythm with agents that sweep every system each morning, surface what needs a decision, and keep nothing waiting silently. Use when running a business day to day and things fall through the cracks between desks.

How it works

  1. Team
  2. Method
  3. Run it
  4. Signals it works
  5. Boundaries
Original SKILL.md headings · 61 lines, 420 words
SKILL.md preview
---
name: agent-ops-command
description: Run the daily operating rhythm with agents that sweep every system each morning, surface what needs a decision, and keep nothing waiting silently. Use when running a business day to day and things fall through the cracks between desks.
---

# Agent operations command

Most operational failure is not a crisis but an omission: an invoice
unsent, a customer unanswered, a deploy half done. A command desk exists
to sweep everything once a day and turn the sweep into a short list a
human can clear, which is the opposite of a dashboard nobody opens.

## Team

- **Sweeper**: checks every system for items past their expected state.
- **Prioritiser** (`prioritize-tasks`): orders findings by consequence
  and deadline rather than by age.
- **Briefer** (`status-updates`): writes the one-page morning brief.

Shape: a scheduled daily sweep producing a single ranked action list.

## Method

1. **Define the sweep list once, explicitly.** Which systems, which
   condition counts as stuck, and what the expected state is. A sweep
   without defined expectations finds nothing.
2. **Look for absence, not just errors.** The invoice not sent and the
   reply not written produce no alert anywhere, and they are most of
   what goes wrong (see agent-company-blueprint).
3. **Rank by consequence and deadline.** A customer waiting two days
   outranks a tidy-up task created this morning, regardless of which was
   raised first.
4. **Cap the brief.** Ten items maximum, with everything else counted
   rather than listed, because an exhaustive brief is one nobody
   finishes reading.
5. **Assign every item an owner and a date.** An item on a list with no
   owner is on the list again tomorrow.
6. **Close the loop the next morning.** Yesterday's items appear with
   their status first, which is what prevents a rolling backlog
   disguised as a daily brief.
7. **Escalate anything aging past a threshold.** Repeated appearance
   without progress is itself the finding.

## Run it

In Claude Code, run the sweeper on a schedule across each desk's output
files and any exported system state, then the prioritiser and briefer in
sequence into a dated brief file. The human clears the list. Port to
CrewAI as a scheduled sequential crew, or LangGraph with a node per
system feeding one ranking node.

## Signals it works

- The brief opens with yesterday's items and their outcomes.
- Items have owners and dates, and aging items escalate.
- Things that quietly stalled are caught within a day rather than a week.

## Boundaries

This desk finds and ranks; it does not act on the items, and anything
touching money, customers, or public channels stops at a human. It
reports on systems it can read, so a system outside the sweep list is
invisible to it. Judgement about what actually matters this week remains
the operator's.

Versions

The database has no other skills named agent-ops-command.

Database snapshot: 13.11.2026 · package version 8fd427e84ba4 (SHA-256 digest of the entire package)

Downloads in the last 30 days: 0 · ranking

Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.

Similar skills

  • 90company-ops
    Use this skill to run affiliate-mcp's day-to-day company operations on a schedule in prepare-and-approve mode: gather everything waiting for the operator's approval (across GitHub, Buffer, and Gmail), collect new inbound feedback and headline metrics, and assemble one daily…
  • 90monday-morning-brief
    Produce the weekly operating brief: what moved, what is at risk, and the three things that need a decision. Use when someone wants a Monday update, a weekly business summary, or a recurring brief across their tools.
  • 95agent-workflow
    Plan and improve day-to-day AI coding-agent workflow across Claude Code, Codex, Gemini CLI, and MCP-heavy repos: session startup, context recovery, fast repo loops, runtime verification, worktree use, and multi-agent handoffs. Use when the user wants a practical operating…
  • 100ops-yolo
    OPS on-demand: This skill should be used when the user asks to \"yolo mode\", \"run the business today\"…

Feedback and reports

Your message goes to Bartosz Sokoliński. An email address is needed only for a reply. See the privacy policy.