agentleFS
koinodkoinod/skills/autonomous-operator/SKILL.md

autonomous-operator

Skill0 starsChanged 6 months agoMITView on GitHub

Provides autonomous business operation capabilities: multi-backend email, local LLM integration, self-monitoring, credential management, and revenue tracking. Enables the agent to run without manual gatekeepers. Provides unified interface: send-email --to <email> --subject <subj> --body <body> Automatically retries with fallback if primary fails. Dashboard can be viewed via: openclaw status --full or a simple web UI. Uses macOS Keychain or encrypted file storage. Environment variables: - AUTONOMOUS_EMAIL_BACKENDS — comma-separated list of backends to try (default: gog,smtp,api,draft) - AUTONOMOUS_LLM_MODEL — local model to…

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every 2 hours: check inbox, send scheduled follow-ups, update lead status

    line 53
  2. Distinctive:

    No dependency on any single credential provider

    line 89

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Autonomous Operator Skill

Provides autonomous business operation capabilities: multi-backend email, local LLM integration, self-monitoring, credential management, and revenue tracking. Enables the agent to run without manual gatekeepers.

Capabilities

  • Multi-backend email — automatically detects and uses available email methods (gog OAuth, SMTP app password, third-party APIs)
  • Local LLM — integrates with Ollama for unlimited tokens, no API costs
  • Self-monitoring — health checks, revenue pipeline tracker, alerting
  • Credential vault — secure storage and rotation of service credentials
  • Autonomous loops — periodic execution of high-value tasks (follow-ups, research, monitoring)

Triggers

  • "Set up autonomous email"
  • "Enable local LLM"
  • "Start self-monitoring"
  • "Build sovereign business machine"
  • "Remove gatekeepers"

What It Does

1. Email Independence

Detects available email sending methods in order of preference:

  1. gog CLI with OAuth (if configured)
  2. SMTP with app password (if credentials available)
  3. Third-party email API (SendGrid, Mailgun, etc.)
  4. Fallback: draft emails for manual send

Provides unified interface: send-email --to <email> --subject <subj> --body <body>

Automatically retries with fallback if primary fails.

2. Local LLM Integration

Installs and manages Ollama on the local machine:

  • Detects if Ollama is installed, installs if not (via brew or direct binary)
  • Pulls Llama 3.2 (or other models) based on available RAM
  • Configures OpenClaw to use local LLM for agent instructions
  • Provides unlimited token generation for agent workflows

3. Self-Monitoring Dashboard

Generates daily status reports:

  • Agent health (running, errors, restarts)
  • Revenue pipeline (leads, emails sent, replies, calls booked, deals closed)
  • Credential status (expiring soon, needs rotation)
  • System resources (RAM, CPU, disk)
  • Upcoming tasks and follow-ups

Dashboard can be viewed via: openclaw status --full or a simple web UI.

4. Autonomous Loops

Runs scheduled tasks without cron (internal scheduler):

  • Every 2 hours: check inbox, send scheduled follow-ups, update lead status
  • Daily: generate revenue report, update next-moves.md, log insights
  • Weekly: competitor watch, pricing intel, template refresh
  • On trigger: when new lead added, automatically enrich and add to sequence

5. Credential Vault

Securely stores:

  • Email passwords/app passwords
  • API keys (OpenRouter, etc.)
  • Client credentials
  • Rotation reminders

Uses macOS Keychain or encrypted file storage.

Setup

# Initialize autonomous operator
openclaw sessions_spawn --runtime subagent --task "Initialize autonomous operator: install Ollama, configure email fallbacks, set up monitoring"

# Or use the CLI directly (once installed)
autonomous-operator init
autonomous-operator email-test
autonomous-operator dashboard

Configuration

Environment variables:

  • AUTONOMOUS_EMAIL_BACKENDS — comma-separated list of backends to try (default: gog,smtp,api,draft)
  • AUTONOMOUS_LLM_MODEL — local model to use (default: llama3.2)
  • AUTONOMOUS_REPORT_TIME — daily report time (default: 09:00)
  • AUTONOMOUS_FOLLOWUP_INTERVAL — hours between follow-up cycles (default: 2)

Benefits

  • No dependency on any single credential provider
  • Works offline (local LLM) for core agent instructions
  • Automatic failover ensures continuity
  • Self-monitoring means you know status at a glance
  • Reduces manual repetitive work to zero

Notes

This skill is designed for sovereignty. It assumes you want to own your infrastructure and not rely on third-party APIs that can go down or change terms. Local LLM gives you unlimited tokens and privacy.

Preferred hardware: Mac mini or Linux server with at least 16GB RAM for comfortable Llama 3.2 operation.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.