Control

What it does

When your agent uses it

• Tasks that involve LLM guardrails
• Tasks that involve PRD writing

Example prompts

• “/control”

Requirements

• Pre-approved tools (allowed-tools): Read, Write, Glob, Grep, Bash, AskUserQuestion

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

1. 1Read runtime control state from .bkit/runtime/control-state.json
2. 2Read trust score from .bkit/state/trust-score.json
3. 3Read active guardrails from lib/control/ configuration
4. 4Display formatted status panel

What it can do on your machine

Read from SKILL.md and the folder at commit 85b4913. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Read
  • Write
  • Glob
  • Grep
  • Bash
  • AskUserQuestion

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are bash).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Control loads about 1.6k tokens when it runs. Until then it costs about 82 tokens; SKILL.md has 437 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~82

When it runs · the whole SKILL.md, loaded when a task matches

~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Read, Write, Glob, Grep, Bash, AskUserQuestion

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from ww-w-ai/bkit-claude-code at commit 85b4913, republished under its Apache-2.0 licence (© ww-w-ai). 437 words, ~1,555 tokens.

name

control

description

Control bkit automation level (L0-L4), view trust score, and manage guardrails. Trust level directly drives SPRINT_AUTORUN_SCOPE (v2.1.13): L0 manual+stopAfter=prd, L4 full-auto+stopAfter=archived. Also gates PDCA phase transitions and destructive operations. Triggers: control, automation level, trust score, guardrail

allowed-tools

Read, Write, Glob, Grep, Bash, AskUserQuestion

classification

workflow

classification-reason

Automation control persists regardless of model advancement

deprecation-risk

none

effort

medium

argument-hint

[status|level|pause|resume|trust]

user-invocable

true

task-template

[Control] {action}

--- bkit Control Panel ----------------------------
Automation Level : L2 (Semi-Auto)
Trust Score      : 72/100
Active Guardrails: 8/8
Paused           : No
Features Active  : 2/3
---------------------------------------------------
Guardrails:
  [ON] Destructive operation detection
  [ON] Blast radius limiter (max 10 files)
  [ON] Loop breaker (max 5 iterations)
  [ON] Checkpoint auto-creation
  [ON] Permission escalation gate
  [ON] Stale feature timeout (7d)
  [ON] Context overflow protection
  [ON] Concurrent write lock
---------------------------------------------------

--- Trust Score Details ---------------------------
Overall Score: 72/100

Breakdown:
  PDCA Completion Rate  : 18/25  (72% cycles completed)
  Match Rate Average    : 22/25  (88% average)
  Error Recovery Rate   : 14/20  (70% recovered)
  Session Stability     : 12/15  (80% stable)
  User Override Freq    :  6/15  (40% override rate)
---------------------------------------------------
Level Recommendation: L2 (Semi-Auto)
  Score 0-30  -> L0 (Manual)
  Score 31-50 -> L1 (Guided)
  Score 51-70 -> L2 (Semi-Auto)
  Score 71-85 -> L3 (Auto)
  Score 86+   -> L4 (Full-Auto)
---------------------------------------------------

# Check current status
/control

# Set to Semi-Auto
/control level 2

# Pause all automation
/control pause

# Resume previous level
/control resume

# View trust score
/control trust

© ww-w-ai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/control of ww-w-ai/bkit-claude-code.

Open the folder on GitHubat commit 85b4913