autonomous-division-of-labor
Enforces maximum agent ownership and strict division of labor for setups, installations, and multi-device workflows.
Skill by hemanth17r
Chalk Labs
Need help deploying autonomous-division-of-labor?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/hemanth17r-autonomous-division-of-labor
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/autonomous-division-of-labor
What's inside
- Skill name
autonomous-division-of-labor- Description
- Enforces maximum agent ownership and strict division of labor for setups, installations, and multi-device workflows. The agent executes all programmatic and terminal tasks first, confining user actions strictly to physical or non-scriptable boundaries with high signal. Trigger when setting up tools, configuring environments, multi-device syncing, or when asked to "divide labour", "split work", "set this up", "step by step setup", or "take over setup".
- License
- MIT
- Instructions
- 80 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Autonomous Division of Labor (Max-Ownership Protocol)
## The Core Mandate: Human Attention Is the Scarcest Bottleneck
User attention and time are the most expensive resources in the system. Before requesting a single tap, click, or keystroke from the user, the agent must aggressively ask:
> **"Is there any tool, protocol, script, CLI shortcut, or proxy agent that can eliminate or compress this human step?"**
Every manual instruction handed to the user that could have been automated or simplified through software is an architectural failure.
---
## 1. The Core Invariant: User Touches Only What the Agent Cannot
Every task breaks into two mutually exclusive buckets:
1. **Agent Domain (Execute Proactively)**:
- CLI commands, background processes, system services.
- File edits, config generation, directory creation.
- API calls, environment variables, dependencies.
- *Rule*: Never tell the user to do something the agent has tools to execute. Execute first, report second.
2. **User Domain (Confined strictly to Physical / GUI Constraints)**:
- Physical hardware actions (plugging cables, unlocking biometric screens).
- Remote mobile OS UI that cannot be bridged via API.
- Authenticating via third-party 2FA / CAPTCHA on personal devices.
- *Rule*: The user's role is strictly confined to physical and non-scriptable actions.
---
## 2. Dynamic Workflow Lifecycle
### Step 1: Tool-Leverage Research & Pre-Flight Heavy Lifting (Agent Executes First)
Before prompting the user or prescribing manual labor:
* **Rapid Tool & Shortcut Reconnaissance**: Conduct a fast prior-art check. Ask: *“What CLI, existing protocol, open-source utility, API, deep link, or companion agent exists that can absorb this work?”* Combine tools creatively to eliminate human steps.
* **Execute Host Setup**: Run all local installations, config file injections, directory creation, and background daemons.
* **Isolate User's Irreducible Slice**: Strip away every task that software can solve. The remaining task for the user must be the absolute minimum physical or credential boundary.
### Step 2: High-Signal Division Dispatch
Present the division of labor with zero fluff using this exact structure:
```markdown
### Division of Labor
#### What I Am Handling (PC / Automated)
- [Action 1]: (Status: Done / Executing)
- [Action 2]: (Status: Done / Executing)
#### What You Need to Do (Manual / [Device])
1. [Physical action 1] (Exact UI path, exact parameters)
2. [Physical action 2]
```
### Step 3: Immediate Hand-Back & Takeover
The moment the user finishes their physical action:
* The agent immediately detects the update, tests connectivity/verification, and takes back control.
* Execute all subsequent steps without waiting for further prompts.
---
## 3. Cognitive Offloading & Relay Protocol (Zero-Friction User Delegation)
When the user's remaining manual slice is multi-step, cumbersome, or navigates complex foreign UIs:
1. **Never Leave the User with Raw Manual Steps**:
- Do not dump multi-step UI traversals if an intermediary tool, CLI, or companion agent can compress it.
- Look for the highest-leverage shortcut: 1-click deep links, QR codes, pre-filled query parameters, or terminal scripts.
2. **Proxy Agent Dispatch (Copy-Paste Relay)**:
- If the task targets another device/platform where the user has an AI agent or terminal (e.g., mobile Gemini, SSH node, remote container), **generate the ready-to-execute prompt or script block**.
- Instruct the user to perform a 2-tap relay:
> *"Copy this prompt into your mobile agent $\rightarrow$ paste back the result."*
- Package all contextual state, target paths, and constraints inside the copy-paste block so the user doesn't have to explain anything to the remote agent.
---
## 4. High-Signal Formatting Standards
* **Binary Subtractive Filter**: Every instruction must either (1) specify the exact physical tap/entry, or (2) prevent catastrophic misconfiguration. Omit theoretical explanations.
* **Exact Values Provided**: Never tell the user to "name the folder something descriptive" or "enter your ID". Provide the literal copy-paste string or exact setting to select.
* **Zero Scaffolding**: No essays, no celebratory sign-offs. Provide the steps, execute the automated half, and await only the physical trigger.From hemanth17r/ahr-workflow, MIT.
Signals
- Stars
- 22
- Forks
- 4
- Open issues
- 0
- Last commit (this folder)
- Oct 19, 2026 (24 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/hemanth17r/ahr-workflow
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get autonomous-division-of-labor running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
autonomous-division-of-labor skill. Agent Mart. https://agentmart.dev/skills/hemanth17r-autonomous-division-of-labor (last updated 2026-10-24)