Rulespec

rulespec

Manage business rules for AI agents without breaking what already works.

Note for OpenClaw users: if rulespec is installed as an OpenClaw plugin (openclaw plugins install rulespec), prefer the native tools (rulespec_init, rulespec_add_rule, rulespec_emit, …) over the CLI — they run in-process with structured params. The plugin variant of this skill at ~/.openclaw/plugin-skills/rulespec/SKILL.md documents them. The CLI docs below apply to every other context (standalone npm, Claude Code, Cursor, Codex, etc.).

Adding a rule to a system prompt shouldn't risk invalidating the ones that are already there. Inline prompt editing doesn't scale — and other solutions aren't built for business rules.

rulespec treats each rule as an independent, validated unit. Add, edit, or remove one rule via CLI — the rest stay untouched. The output is a structured SKILL.md that any AI agent can load.

IMPORTANT: Always use the rulespec CLI to modify rules, sources, and examples. Never edit emitted SKILL.md files directly — they are generated and will be overwritten. For complex structures (source schemas, nested example data), you may edit the source rulespec.yaml directly, but always run rulespec validate afterward.

All commands use npx rulespec — no global install needed. npx downloads and runs it automatically.

Layout

Each domain lives in its own skill folder. By default it lands under skills/; with -a <agent> it lands in that agent's skills directory instead:

skills/                       ← default
  invoice-processing/
    rulespec.yaml   ← authored source
    SKILL.md        ← emitted, agent-loadable (do not edit)

.claude/skills/               ← rulespec init -a claude-code
  customer-support/
    rulespec.yaml
    SKILL.md

CLI commands

Setup

rulespec init --domain "invoice processing"                  # → skills/invoice-processing/
rulespec init --domain "invoice processing" -a claude-code   # → .claude/skills/invoice-processing/
rulespec init --domain "invoice processing" -a claude-code -g # → ~/.claude/skills/invoice-processing/
rulespec set-domain "customer support"                       # Change the domain

Supported agents for -a: claude-code, cursor, openclaw, codex, opencode, windsurf, amp, augment, gemini-cli, github-copilot. Anywhere else: pass --outdir <path>.

Once set at init, every other command auto-discovers the source — no need to repeat -a.

Rules

rulespec add --id <id> --rule <text> --context <text> --intent <enforce|inform|suggest>
rulespec edit <id> --rule <new text>           # Update rule text
rulespec edit <id> --intent enforce            # Change intent level
rulespec edit <id> --context "new context"     # Change when rule applies
rulespec remove <id>                           # Remove a rule
rulespec list                                  # List all rules

Sources

rulespec add-source --id <id> --type <document|api|database|message|structured> --description <text> [--format <fmt>]
rulespec remove-source <id>

Global examples (end-to-end input/output pairs)

rulespec add-example --input '{"key": "val"}' --output '{"key": "val"}' [--note <text>]
rulespec add-example --input /path/to/input.json --output /path/to/output.json --note "From files"
rulespec add-example --input /path/to/invoice.pdf --output '{"action": "approve"}' --note "PDF input"
rulespec remove-example <index>                # 0-based index

Rule-specific examples

rulespec add-rule-example <rule-id> --input '{"amount": 100}' --output '{"approved": true}'
rulespec add-rule-example <rule-id> --input /path/to/file.pdf --output '{"extracted": "data"}'
rulespec remove-rule-example <rule-id> <index> # 0-based index

Input/output resolution

Both --input and --output accept three formats:

• Inline JSON: '{"key": "val"}' — parsed directly
• JSON file path: /path/to/data.json — file is read and parsed
• Any other file path: /path/to/doc.pdf — stored as { file: "/path/to/doc.pdf" }

Find & replace

rulespec replace --old "30 days" --new "60 days"   # Validates + recompiles automatically

Build & emit

rulespec compile [id]                          # Preview compiled prompts
rulespec validate                              # Check file against schema
rulespec emit                                  # Generate skills/{domain}/SKILL.md
rulespec emit --include-examples true          # Include examples in output
rulespec emit --outdir <path>                  # Custom output dir (default: skills)

All commands accept --file <path> to specify a different file (auto-detected from skills/*/rulespec.yaml).

File format

schema: rulespec/v1
domain: "your domain here"

sources:                              # optional — what data the rules operate on
  - id: source-name
    type: document | api | database | message | structured
    format: pdf | json | csv          # optional
    description: "What this source is"
    schema:                           # optional — shape of the data
      field: type

rules:
  - id: rule-id                       # kebab-case, unique
    rule: "The business rule in plain language"
    context: "When this rule applies"
    intent: enforce | inform | suggest

examples:                             # optional — end-to-end golden standards
  - note: "What this example tests"
    input: { ... }
    output: { ... }

Intent levels

• enforce — mandatory. Agent must follow this rule. Compiles to directive language.
• inform — guidance. Agent should be aware. Compiles to neutral language.
• suggest — recommendation. Agent may consider. Compiles to soft language.

Workflow

1. rulespec init --domain "my domain" scaffolds skills/{domain}/rulespec.yaml
2. Add sources with rulespec add-source
3. Add rules with rulespec add
4. Add examples with rulespec add-example
5. rulespec validate to check for errors
6. rulespec compile to preview compiled prompts
7. rulespec emit to generate skills/{domain}/SKILL.md next to the source

Programmatic usage

To inject rules into LLM prompts at runtime:

import { loadRules } from "rulespec";
const rules = await loadRules("skills/my-domain/rulespec.yaml");
// rules is a compiled markdown string — inject into any system prompt or API call

Key principles

• Use the CLI for rules, sources, and examples — never edit emitted SKILL.md files
• For complex source schemas or nested example data, edit the source rulespec.yaml directly + run rulespec validate
• rulespec replace is a safe find-and-replace: validates and recompiles after every change
• One rule, one change — editing a rule only affects that rule's compiled output
• Examples are excluded from emitted SKILL.md by default (they may contain sensitive data)

Built by the team behind Clawnify.