Autonomy
Verified@ivangdavila
npx machina-cli add skill @ivangdavila/autonomy --openclawData Storage
~/autonomy/
├── tracking.md # What's been delegated, success rates
├── proposals.md # Pending takeover proposals
└── rejected.md # User declined, don't re-propose
Create on first use: mkdir -p ~/autonomy
Scope
This skill:
- ✅ Identifies repetitive tasks from conversation history
- ✅ Proposes delegation opportunities to user
- ✅ Tracks success rate of delegated tasks
- ❌ NEVER acts autonomously without explicit prior approval
- ❌ NEVER observes outside of conversation context
- ❌ NEVER accesses files/systems to "audit" user activity
- ❌ NEVER monitors calendar/email without permission
Quick Reference
| Topic | File |
|---|---|
| Bottleneck detection | bottlenecks.md |
| Takeover process | expansion.md |
Core Rules
1. Learning Source
Identify delegation candidates ONLY from:
- Explicit user statements ("I always have to do X")
- Repeated requests in conversation ("deploy again", "same as before")
- User complaints about repetitive work
NEVER from:
- Accessing user's calendar/email to find patterns
- Monitoring file changes or system activity
- Any form of surveillance
2. Bottleneck Signals (conversation-based)
| Signal | Example |
|---|---|
| Repeated request | "Deploy to staging" every PR |
| Rubber-stamp | User always approves without changes |
| Complaint | "I hate doing this every time" |
3. Takeover Proposal
When you spot a pattern in conversation:
💡 Delegation opportunity
I noticed: [what you observed in our chats]
Pattern: [how often you've asked for this]
Proposal: I could handle [specific task] without asking each time.
Pilot: First 5x I'll do it and tell you after.
Then: Full autonomy if you're happy.
Want to try?
4. Expansion Levels
| Level | Description |
|---|---|
| L1 | Do what's asked |
| L2 | Fill gaps, handle edge cases |
| L3 | Own workflows after pilot approval |
Always requires explicit user approval to move up levels.
5. Tracking
In ~/autonomy/tracking.md:
## Delegated
- deploy/staging: approved 2024-01, 50+ successful
- code-review/style: approved 2024-02, 200+ runs
## Pilot Phase
- deploy/production: 3/5 runs, pending full approval
6. Anti-Patterns
| Don't | Do instead |
|---|---|
| Take over without asking | Always propose first |
| Monitor user activity | Only observe conversations |
| Assume after one approval | Confirm scope each time |
Overview
Autonomy expands agent capabilities by spotting repetitive tasks within conversation history, proposing delegation opportunities, and tracking success rates. It stores data under ~/autonomy and enforces strict boundaries: no autonomous actions, no outside surveillance, and no calendar or file access without explicit permission.
How This Skill Works
The system identifies delegation candidates from explicit user statements, repeated requests, or complaints in chat. When a pattern is detected, it presents a takeover proposal and initiates a pilot (first 5x) before any full autonomy, recording results in ~/autonomy/tracking.md. It never acts without explicit user approval and limits operations to the current conversation context, with expansion progressing only via explicit consent (L1-L3).
When to Use It
- When a task is repeatedly requested in conversations (e.g., 'deploy to staging' every PR).
- When a user states they always have to do a task (e.g., 'I always do X').
- When a user complains about repetitive work (e.g., 'I hate doing this every time').
- When you want to pilot delegation with a 'first 5x' approach before full autonomy.
- When you want to track performance and decide on moving from L1 to L3 with explicit consent.
Quick Start
- Step 1: Identify delegate-worthy tasks from chat history.
- Step 2: Present the takeover proposal and awaits explicit approval.
- Step 3: If approved, run the pilot (first 5x) and log results in ~/autonomy/tracking.md.
Best Practices
- Propose delegation first using the Takeover Proposal template before acting.
- Constrain actions to data within the chat context; never access calendars, emails, or files without permission.
- Run a pilot (first 5x) and report results before expanding scope.
- Explicitly confirm the scope and level (L1-L3) of autonomy for each task.
- Document outcomes in ~/autonomy/tracking.md and iterate based on success.
Example Use Cases
- Detects repeated 'deploy to staging' requests and proposes delegation after a 5x pilot.
- Triggers auto-review of standard formatting across PRs only after user approval.
- Notes a user complaint about repetitive tasks and offers to handle them with explicit consent.
- Keeps a running tally of delegated tasks and success rates in ~/autonomy/tracking.md.
- Escalates to full autonomy (L3) only after the pilot proves reliable with explicit approval.
