Skip to content
Get packs

Agent-Operated Workflows

Teaches an agent to run bounded, multi-step procedures by observing, verifying, and checkpointing its own progress

Source Download skill
Repository
mvschwarz/openrig
Stars
8,553
Licence
Apache-2.0
Files
1 file

What it does

This skill gives an AI agent a design philosophy for carrying out a bounded real-world procedure whose exact steps depend on messy, partly unknown current conditions. Instead of replaying a fixed script, the agent inspects state with read-only checks, picks one safe action, verifies the result, records a checkpoint so another agent can resume, and escalates to a human when something is genuinely ambiguous or destructive. It also helps decide whether a task should be run by agent judgment or by ordinary deterministic code.

Great for

  • Developers designing automated operations or runbooks that must survive partial failures and be resumable
  • Teams running upgrades or maintenance on existing (brownfield) systems where state varies
  • People building agent-driven workflows who need guidance on where judgment versus code should sit

Skip it if

  • Simple, fully predictable tasks like a parser or fixed data transformation, where the skill itself says ordinary code is better
  • Fully autonomous destructive operations: the skill says humans own destructive ambiguity, so an agent following it should still stop and ask in those cases

Try asking

  • “Use the agent-operated-workflows approach to plan how our nightly database migration should inspect state, checkpoint, and escalate”
  • “Help me decide whether this customer-data cleanup should be an agent loop or a plain script”
  • “Walk through this multi-step server upgrade and verify each step before moving on, stopping if anything is unclear”
Step-by-step guide: how to use Agent-Operated WorkflowsInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the Apache-2.0 licence and an attribution note.

    agent-operated-workflows.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip agent-operated-workflows.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip agent-operated-workflows.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Point the agent at a concrete procedure and its runbook; the skill is a contract and philosophy, not a ready-made tool
  • The skill's public example relies on OpenRig's openrig-upgrade runbook and rig context commands, which are only available inside the OpenRig environment
  • Expect the skill to be marked provisional by its authors; treat its guidance as an evolving reference rather than settled standard

What's inside

  • SKILL.md

Redistributed unmodified from mvschwarz/openrig under the Apache-2.0 licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.