Control Skill
View and manage bkit automation levels and guardrail settings
Commands
| Command |
Description |
Example |
/control status |
Show current automation state |
/control status |
/control level L0 |
Set automation to manual |
/control level L0 |
/control level L1 |
Set automation to semi-auto |
/control level L1 |
/control level L2 |
Set automation to auto-confirm |
/control level L2 |
/control level L3 |
Set automation to full-auto (read-only) |
/control level L3 |
/control level L4 |
Set automation to autonomous |
/control level L4 |
/control stop |
Emergency stop (immediate L0 + halt) |
/control stop |
/control rollback |
Rollback last N decisions |
/control rollback 5 |
Automation Levels (v2.0.7+ L3/L4 unlocked via bkit hook automation channel)
| Level |
Name |
Behavior |
| L0 |
Manual |
Agent asks before every action. No automatic file writes. Emergency-stop default. |
| L1 |
Semi-Auto |
Agent executes known-safe operations. Asks for destructive actions. |
| L2 |
Auto-Confirm |
Agent executes all operations. Confirms only for irreversible actions. Default for new bkit projects. |
| L3 |
Full-Auto |
Read-only tools (Read/Glob/Grep/list_directory/web_fetch) auto-allowed via bkit before-tool hook. Destructive ops always asked. Requires trustScore >= 60. |
| L4 |
Autonomous |
All non-destructive tools auto-allowed. PDCA phase transitions automatic (via before-phase hook). Destructive ops always hard-denied (D6, 16 patterns matrix). Requires trustScore >= 80 + no recent rejections + no active 24h cooldown. |
Hard Deny (영구 차단, L4 무관)
다음 16 destructive 패턴은 어떤 level에서도 항상 차단됩니다 (cmd-parser AST 분석):
rm -rf, git push --force, drop table, truncate, delete from (no WHERE),
sudo, chmod 777, /etc/* write, ~/.ssh/* write, gpg --delete-key,
.env write, id_rsa* / *.pem / *.p12 write/delete, npm publish --force,
git config --unset credential.*, rm .git, drop database.
Auto-Downgrade (안전망)
- 5 consecutive rejections → L4/L3 → L2 + 24h cooldown
- Hard deny detected (destructive op) → L4 → L3 (no cooldown)
- Emergency stop (Ctrl+C 2회 또는
/control stop) → 즉시 L0
Gemini CLI 호환성 (v2.0.7 Wave 0 RCA로 입증)
- Non-interactive
gemini -p 모드: 'allow' decision suppress 정상 작동 ✓
- Interactive mode: CC issue #52822와 유사 가능성, Track B PR 검증 중
- 모든 결정은
.bkit/state/audit/{date}/decisions.jsonl에 영구 기록 (90d retention)
How to Execute
Show Status
- Read
.bkit/config.json for current settings
- Display:
- Current automation level (L0-L2)
- Active feature flags
- Trust score (if tracked)
- Hook configuration summary
- Guardrail rules in effect
Change Level
- Read
.bkit/config.json
- Validate the requested level (L0, L1, or L2 only)
- Update the
automationLevel field
- Write back to
.bkit/config.json using write_file
- Confirm the change and explain what it means
Configuration File
Settings are stored in .bkit/config.json:
{
"automationLevel": "L1",
"features": {
"auditLog": true,
"checkpointOnDeploy": true,
"autoIterate": false,
"maxIterations": 5
},
"guardrails": {
"denyPatterns": [".env", "credentials", "secret"],
"maxFileSize": 10000,
"requireCheckpoint": true
}
}
Feature Flags
| Flag |
Default |
Description |
auditLog |
true |
Enable audit trail logging |
checkpointOnDeploy |
true |
Auto-create checkpoint before deploy |
autoIterate |
false |
Auto-run iterate when match rate < 90% |
maxIterations |
5 |
Maximum PDCA iteration cycles |
Guardrail Rules
Guardrails are enforced by Extension hooks (before-model, before-agent):
- File deny patterns: Block writes to sensitive files
- Max file size: Prevent accidentally large file writes
- Checkpoint requirement: Require checkpoint before destructive operations
- Review requirement: Require human review for production changes
Output Format
## bkit Control Status
| Setting | Value |
|---------|-------|
| Automation Level | L1 (Semi-Auto) |
| Audit Logging | Enabled |
| Checkpoint on Deploy | Enabled |
| Auto-Iterate | Disabled |
| Max Iterations | 5 |
### Active Guardrails
- File deny patterns: .env, credentials, secret
- Max file size: 10,000 lines
- Checkpoint required: Yes
1---2name: control3description: View and change bkit automation settings. Shows current automation level, feature flags, trust score, and guardrail configuration. Use proactively when user asks about automation settings, trust level, or guardrails. Triggers: control, automation, settings, level, guardrail, trust, 제어, 자동화, 설정, 레벨, 가드레일, 신뢰도, 制御, 自動化, 設定, レベル, ガードレール, 控制, 自动化, 设置, 级别, 护栏, control, automatización, configuración, nivel, contrôle, automatisation, paramètres, niveau, Kontrolle, Automatisierung, Einstellungen, Stufe, controllo, automazione, impostazioni, livello Do NOT use for: deployment (use /deploy), code changes4---56# Control Skill78> View and manage bkit automation levels and guardrail settings910## Commands1112| Command | Description | Example |13|---------|-------------|---------|14| `/control status` | Show current automation state | `/control status` |15| `/control level L0` | Set automation to manual | `/control level L0` |16| `/control level L1` | Set automation to semi-auto | `/control level L1` |17| `/control level L2` | Set automation to auto-confirm | `/control level L2` |18| `/control level L3` | Set automation to full-auto (read-only) | `/control level L3` |19| `/control level L4` | Set automation to autonomous | `/control level L4` |20| `/control stop` | Emergency stop (immediate L0 + halt) | `/control stop` |21| `/control rollback` | Rollback last N decisions | `/control rollback 5` |2223## Automation Levels (v2.0.7+ L3/L4 unlocked via bkit hook automation channel)2425| Level | Name | Behavior |26|-------|------|----------|27| **L0** | Manual | Agent asks before every action. No automatic file writes. Emergency-stop default. |28| **L1** | Semi-Auto | Agent executes known-safe operations. Asks for destructive actions. |29| **L2** | Auto-Confirm | Agent executes all operations. Confirms only for irreversible actions. Default for new bkit projects. |30| **L3** | Full-Auto | Read-only tools (Read/Glob/Grep/list_directory/web_fetch) auto-allowed via bkit before-tool hook. Destructive ops always asked. Requires `trustScore >= 60`. |31| **L4** | Autonomous | All non-destructive tools auto-allowed. PDCA phase transitions automatic (via before-phase hook). Destructive ops always **hard-denied** (D6, 16 patterns matrix). Requires `trustScore >= 80 + no recent rejections + no active 24h cooldown`. |3233### Hard Deny (영구 차단, L4 무관)3435다음 16 destructive 패턴은 어떤 level에서도 항상 차단됩니다 (cmd-parser AST 분석):36`rm -rf`, `git push --force`, `drop table`, `truncate`, `delete from (no WHERE)`,37`sudo`, `chmod 777`, `/etc/* write`, `~/.ssh/* write`, `gpg --delete-key`,38`.env write`, `id_rsa* / *.pem / *.p12 write/delete`, `npm publish --force`,39`git config --unset credential.*`, `rm .git`, `drop database`.4041### Auto-Downgrade (안전망)4243- **5 consecutive rejections** → L4/L3 → L2 + 24h cooldown44- **Hard deny detected** (destructive op) → L4 → L3 (no cooldown)45- **Emergency stop** (Ctrl+C 2회 또는 `/control stop`) → 즉시 L04647### Gemini CLI 호환성 (v2.0.7 Wave 0 RCA로 입증)4849- Non-interactive `gemini -p` 모드: `'allow'` decision suppress 정상 작동 ✓50- Interactive mode: CC issue #52822와 유사 가능성, Track B PR 검증 중51- 모든 결정은 `.bkit/state/audit/{date}/decisions.jsonl`에 영구 기록 (90d retention)5253## How to Execute5455### Show Status56571. Read `.bkit/config.json` for current settings582. Display:59 - Current automation level (L0-L2)60 - Active feature flags61 - Trust score (if tracked)62 - Hook configuration summary63 - Guardrail rules in effect6465### Change Level66671. Read `.bkit/config.json`682. Validate the requested level (L0, L1, or L2 only)693. Update the `automationLevel` field704. Write back to `.bkit/config.json` using `write_file`715. Confirm the change and explain what it means7273## Configuration File7475Settings are stored in `.bkit/config.json`:7677```json78{79 "automationLevel": "L1",80 "features": {81 "auditLog": true,82 "checkpointOnDeploy": true,83 "autoIterate": false,84 "maxIterations": 585 },86 "guardrails": {87 "denyPatterns": [".env", "credentials", "secret"],88 "maxFileSize": 10000,89 "requireCheckpoint": true90 }91}92```9394## Feature Flags9596| Flag | Default | Description |97|------|---------|-------------|98| `auditLog` | `true` | Enable audit trail logging |99| `checkpointOnDeploy` | `true` | Auto-create checkpoint before deploy |100| `autoIterate` | `false` | Auto-run iterate when match rate < 90% |101| `maxIterations` | `5` | Maximum PDCA iteration cycles |102103## Guardrail Rules104105Guardrails are enforced by Extension hooks (before-model, before-agent):1061071. **File deny patterns**: Block writes to sensitive files1082. **Max file size**: Prevent accidentally large file writes1093. **Checkpoint requirement**: Require checkpoint before destructive operations1104. **Review requirement**: Require human review for production changes111112## Output Format113114```markdown115## bkit Control Status116117| Setting | Value |118|---------|-------|119| Automation Level | L1 (Semi-Auto) |120| Audit Logging | Enabled |121| Checkpoint on Deploy | Enabled |122| Auto-Iterate | Disabled |123| Max Iterations | 5 |124125### Active Guardrails126- File deny patterns: .env, credentials, secret127- Max file size: 10,000 lines128- Checkpoint required: Yes129```