rule-creation

# rule-creation Skill ## Description Auto-triggers **Context Engineering** compliance and Lobster enforcement evaluation when **proposing, creating, or formalizing new rules, policies, processes, workflows, or behavioral constraints**. **Context Engineering** is the practice of wiring rules into your AI's context so they're reliably followed — not just hoped for via prompting. Credit: [Muratcan Koylan (@koylanai)](https://x.com/koylanai/status/2025286163641118915) Use this skill any time you are: - Declaring a new rule (e.g. "From now on, always X") - Formalizing a policy (e.g. "The policy for Y is Z") - Creating a behavioral constraint (e.g. "Never do X without doing Y first") - Proposing a process or workflow - Updating governance documents (TOOLS.md, CRITICAL-RULES.md, CONSTITUTION.md, etc.) --- ## When This Skill Matches **Trigger phrases:** - "Add this rule…" - "Create a policy…" - "From now on…" - "The rule is…" - "New constraint…" - "Formalize that…" - "Wire this rule…" - "Make it a HARD rule…" - Any governance/policy update request --- ## Mandatory Workflow When this skill triggers, **always** execute the following steps in order. Do NOT ask permission — CE-PROCESS compliance means **wire first, report after**. ### Step 1: Identify the Rule Extract: - **Rule name** — short slug (e.g. `no-bare-spawns`, `external-message-gate`) - **Rule description** — plain language statement of what the rule requires - **Rule type** — `HARD` (non-negotiable, always enforced) or `soft` (guideline, use judgment) ### Step 2: Evaluate Enforcement Need (CR-012 Criteria) Run `scripts/evaluate-enforcement-need.js` or apply manually: A rule **needs Lobster enforcement** if ANY of the following are true: | Criterion | Examples | |-----------|----------| | **High-stakes** | Sends external messages, spends money, deletes data, publishes public content, deploys to production | | **Pattern of violations** | A similar rule was already stated but broken or ignored | | **Multi-step sequence** | Requires multiple ordered steps (do A, then B, then C) | | **External / public action** | Affects things outside the user's local files | **Scoring:** - 0 criteria matched → `needsEnforcement: false` (wire to TOOLS.md only) - 1+ criteria matched → `needsEnforcement: true` (create Lobster workflow + wire) ### Step 3: Lobster Availability Preflight Before creating any workflow, verify Lobster is actually enabled: Run `scripts/check-lobster-available.js` **or** check manually: ```bash openclaw plugins list | grep -i lobster ``` | Output | Meaning | Action | |--------|---------|--------| | `lobster … loaded` | ✅ Available | Proceed to Step 3a | | `lobster … disabled` | ❌ Unavailable | Use fallback (below) | | No lobster row | ❌ Not installed | Use fallback (below) | **Fallback when Lobster is unavailable:** - Skip workflow creation entirely - Proceed directly to Step 4 (wire to docs) — no Lobster reference in the entry - Report: `⚠️ Enforcement: unavailable (Lobster not enabled)` - Include hint: `Run: openclaw plugins enable lobster` to activate **Never claim enforcement is wired when Lobster is disabled.** ### Step 3a: Create Lobster Workflow (enforcement needed AND Lobster available) If `needsEnforcement: true` **and** Lobster preflight passed: 1. Copy `templates/lobster-workflow.template.lobster` 2. Fill in: `name`, `description`, steps specific to rule enforcement 3. Save to `workflows/<rule-name>.lobster` 4. Note the workflow path for wiring ### Step 4: Wire to Docs Run `scripts/wire-rule-to-docs.js` or manually append to the appropriate file: **Routing:** - Operational rules → `TOOLS.md` - Identity/persona constraints → `SOUL.md` or `agents/identity.md` - Hard system rules → `CRITICAL-RULES.md` (if exists) or `AGENTS.md` - Project-specific rules → relevant project board or `agents/projects.md` **Entry format** (from `templates/tools-md-entry.template.md`): ``` ## <Rule Name> (<HARD|soft>) <Rule description> [If Lobster enforcement]: **Enforced by:** `workflows/<rule-name>.lobster` ``` ### Step 5: Report What Was Done Your response must include: ``` ✅ Rule wired: <Rule Name> 📄 Location: <file where it was added> 🔒 Enforcement: <Lobster workflow path OR "TOOLS.md entry only"> 📋 Type: <HARD|soft> ``` **Never say:** - ❌ "Want me to add this to TOOLS.md?" - ❌ "Should I wire this through CE?" - ❌ "Is this CE compliant?" **Always say:** - ✅ "Added `<rule>` to `<file>`. Enforcement: `<status>`." --- ## Script Reference | Script | Purpose | |--------|---------| | `scripts/check-lobster-available.js` | Preflight: verify Lobster plugin is enabled in this OpenClaw instance | | `scripts/evaluate-enforcement-need.js` | Evaluate CR-012 criteria against rule description | | `scripts/wire-rule-to-docs.js` | Append rule entry to TOOLS.md or specified target file | | `scripts/report-rule-creation.js` | Generate user-facing summary of what was created | --- ## Workflow Reference | Workflow | Purpose | |----------|---------| | `workflows/create-rule.lobster` | Full end-to-end rule creation pipeline | --- ## Context Engineering Auto-Trigger (HARD) This skill embodies the Context Engineering Auto-Trigger principle: > "When proposing any new rule, behavioral constraint, or policy change: Wire it first. Add the rule to the appropriate file BEFORE responding. Never ask permission to formalize a rule." The act of recognizing a rule IS the trigger. Wire immediately. Report after.