Skip to content
Agentmart
Submit a tool

autonomy

Activate when a subagent completes work and needs continuation check.

Skill by intelligentcode-ai

IndexedMIT★ 1
Write a review

Chalk Labs

Need help deploying autonomy?

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.

Book a call

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/intelligentcode-ai-autonomy

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.

We use your email only for this newsletter. Privacy

About

Activate when a subagent completes work and needs continuation check. Activate when a task finishes to determine next steps or when detecting work patterns in user messages. Governs automatic work continuation and queue management.

Use cases: productivity, agent-building

No 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. 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/autonomy

What's inside

Skill name
autonomy
Description
Activate when a subagent completes work and needs continuation check. Activate when a task finishes to determine next steps or when detecting work patterns in user messages. Governs automatic work continuation and queue management.
License
MIT
Instructions
317 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Autonomy Skill

**Invoke automatically** after subagent completion or when deciding next actions.

## When to Invoke (Automatic)

| Trigger | Action |
|---------|--------|
| Subagent returns completed work | Check selected tracking backend for next item |
| Task finishes successfully | Update status, pick next pending item |
| Work pattern detected in user message | Create work items if L2/L3 |
| Multiple tasks identified | Queue all, parallelize if L3 |

## Acceptance Tests

| Test ID | Type | Prompt / Condition | Expected Result |
| --- | --- | --- | --- |
| AUT-T1 | Positive trigger | "Continue with autonomy after this item completes" | skill triggers |
| AUT-T2 | Positive trigger | "Run the next queued item automatically" | skill triggers |
| AUT-T3 | Negative trigger | "Explain what autonomy levels mean" | skill does not auto-dispatch work |
| AUT-T4 | Behavior | no system autonomy level in user ica-config | ask user to set system level (`L1`/`L2`/`L3`), persist choice |
| AUT-T5 | Behavior | no project autonomy level in project ica-config | ask user to set project level (`follow-system`/`L1`/`L2`/`L3`), persist choice |
| AUT-T6 | Behavior | project level is `follow-system` | effective level resolves to system level |
| AUT-T7 | Behavior | project level explicitly set | effective level uses project level override |
| AUT-T8 | Behavior | actionable findings arrive while an item is already `in_progress` | MUST invoke `create-work-items` + `plan-work-items`; `run-work-items` records deferred dispatch and does not preempt |
| AUT-T9 | Behavior | any continuation cycle executes | MUST include all three skills (`create-work-items`, `plan-work-items`, `run-work-items`) with invocation evidence or fail-closed |

## Autonomy Level Configuration (MANDATORY)

Autonomy level MUST be resolved from `ica.config.json` (not tracking config).

Level values:
- `L1`
- `L2`
- `L3`

Project-only value:
- `follow-system`

Required keys:
- System (user) config key: `autonomy.system_level` (`L1` | `L2` | `L3`, default `L2`)
- Project config key: `autonomy.project_level` (`follow-system` | `L1` | `L2` | `L3`, default `follow-system`)

Resolution order:
1. Resolve user (system) config file:
   - `${ICA_HOME}/ica.config.json`
   - `$HOME/.codex/ica.config.json`
   - `$HOME/.claude/ica.config.json`
2. Resolve project config file:
   - `./ica.config.json` (project root)
3. Read `autonomy.system_level` from user config.
4. Read `autonomy.project_level` from project config.
5. Determine effective level:
   - if `autonomy.project_level` is `L1`/`L2`/`L3`, use that
   - if `autonomy.project_level` is `follow-system`, use `autonomy.system_level`

Bootstrap prompts (required if missing):
- If `autonomy.system_level` is missing:
  - ask: "No system autonomy level is configured. Set system autonomy level to `L2` (recommended), `L1`, or `L3`?"
  - persist in user `ica.config.json`
- If `autonomy.project_level` is missing:
  - ask: "No project autonomy level is configured. Set project autonomy level to `follow-system` (recommended), `L1`, `L2`, or `L3`?"
  - persist in project `ica.config.json`

Persistence rules:
- System level MUST be written to user `ica.config.json`.
- Project level MUST be written to project `ica.config.json`.
- If both are missing, ask and persist both before auto-dispatch.
- Effective default behavior after bootstrap is `L2` via `system_level=L2` and `project_level=follow-system`.

Example user `ica.config.json`:
```json
{
  "autonomy": {
    "system_level": "L2"
  }
}
```

Example project `ica.config.json`:
```json
{
  "autonomy": {
    "project_level": "follow-system"
  }
}
```

Compatibility:
- If legacy `autonomy.level` exists and `autonomy.system_level` is missing, treat legacy value as system level for this run and persist it as `autonomy.system_level`.

## Tracking Backend Selection (MANDATORY)

Resolve backend with config-first precedence:

1. Project config: `.agent/tracking.config.json`
2. Global config: `${ICA_HOME}/tracking.config.json`
3. Agent-home config fallback:
- `$HOME/.codex/tracking.config.json`
- `$HOME/.claude/tracking.config.json`
4. Auto-detect GitHub backend
5. Fallback: `.agent/queue/`

Provider values:
- `github`
- `linear` (future)
- `jira` (future)
- `file-based`

Detection pattern:

```bash
TRACKING_PROVIDER=""
for c in ".agent/tracking.config.json" \
         "${ICA_HOME:-}/tracking.config.json" \
         "$HOME/.codex/tracking.config.json" \
         "$HOME/.claude/tracking.config.json"; do
  if [ -n "$c" ] && [ -f "$c" ]; then
    TRACKING_PROVIDER="$(python3 - <<'PY' "$c"

From intelligentcode-ai/skills, MIT.

More from intelligentcode-ai/skills

Signals

Stars
1
Forks
1
Open issues
0
Last commit (this folder)
Feb 16, 2026 (8 months ago)
Last release
Feb 16, 2026 (8 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from intelligentcode-ai

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 autonomy 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.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying autonomy?

CallGet help
Cite this page

autonomy skill. Agent Mart. https://agentmart.dev/skills/intelligentcode-ai-autonomy (last updated 2026-10-24)