regsim-in 0.1.0

Regulatory Simulation CLI for Indian Backend Systems

RegSim-IN

Regulatory Simulation & Failure Memory CLI for Indian Backend Systems

---

What is RegSim-IN?

RegSim-IN is a developer-first CLI that simulates Indian regulatory rules (TDS, GST, RBI-style constraints) against backend data flows before production.

It helps backend teams detect, explain, and remember regulatory failures early — during development, testing, and CI — instead of discovering them during audits or incidents.

RegSim-IN treats regulation as executable rules, not PDFs.

---

What Problem Does This Solve?

Backend teams frequently:

• Ship compliant-looking code that fails under real regulatory edge cases
• Discover issues late (audits, settlements, reversals)
• Repeat the same regulatory mistakes across services and teams

RegSim-IN exists to:

• Shift regulatory failures left
• Make rules explicit and testable
• Prevent repeat regulatory incidents

---

What RegSim-IN v1 Does

Version 1 focuses on deterministic rule simulation.

RegSim-IN v1 can:

• Load regulatory rules defined in JSON
• Run those rules against input payloads (also JSON)
• Evaluate pass/fail conditions
• Emit machine-readable JSON output
• Explain why a rule failed

This makes RegSim-IN suitable for:

• Local development checks
• CI/CD gates
• Backend design validation
• Regulatory edge-case exploration

---

Example: Detecting a Missed TDS Deduction

Consider a backend payout flow where a contractor payment is executed without deducting TDS.

Input payload:

{
  "payment": {
    "id": "pay_1029",
    "amount": 45000,
    "vendor_type": "contractor",
    "tds_deducted": false
  }
}

Simulation result:

{
  "status": "FAIL",
  "violations": [
    {
      "rule_id": "TDS_194C_THRESHOLD",
      "severity": "HIGH",
      "message": "TDS must be deducted under section 194C"
    }
  ]
}

This allows teams to catch deduction timing and threshold violations before payouts reach production systems.

---

What RegSim-IN Explicitly Does NOT Do

To avoid misuse or false confidence, RegSim-IN v1 does not:

• Provide legal, tax, or regulatory advice
• File or generate GST / TDS / RBI reports
• Integrate with government, bank, or tax APIs
• Automatically update rules from circulars
• Fully simulate async systems (queues, retries, persistent state)

This is a simulation tool, not a compliance authority.

---

Supported Languages

• Python (v1)

The CLI is language-agnostic, but rule evaluation currently targets Python-style backend data models.

---

Rule Format

Rules are defined in JSON.

Design goals:

• Explicit structure
• Deterministic evaluation
• Easy diffing & review
• CI/CD friendliness

Example Rule

{
  "rule_id": "TDS_194C_THRESHOLD",
  "rule_version": "1.0",
  "effective_from": "2024-04-01",
  "description": "TDS applies if contractor payment exceeds threshold",
  "condition": {
    "field": "payment.amount",
    "operator": ">",
    "value": 30000
  },
  "action": {
    "type": "FAIL",
    "message": "TDS must be deducted under section 194C"
  },
  "source_reference": "Income Tax Act - Section 194C"
}

---

Input Format

Inputs represent backend payloads or traces, also in JSON.

---

Output Format

All outputs are JSON only.

Example failure output:

{
  "status": "FAIL",
  "violations": [
    {
      "rule_id": "TDS_194C_THRESHOLD",
      "severity": "HIGH",
      "message": "TDS must be deducted under section 194C"
    }
  ],
  "metadata": {
    "engine": "regsim-in",
    "engine_version": "0.1.0",
    "rule_snapshot": "2024-04-01",
    "applied_rules": [
      {
        "rule_id": "TDS_194C_THRESHOLD",
        "rule_version": "1.0",
        "effective_from": "2024-04-01",
        "source_reference": "Income Tax Act - Section 194C"
      }
    ]
  }
}

This makes RegSim-IN suitable for automation and tooling.

---

Error Output

Errors are always returned as JSON and never include a Python traceback:

{
  "status": "ERROR",
  "message": "Rule validation failed: missing field 'action'"
}

---

Installation (Early Prototype)

pip install regsim-in

RegSim-IN is under active development.

---

Usage (v1)

regsim-in simulate \
  --rules rules/ \
  --input input.json

Current v1 behavior:

• CLI initializes correctly
• Rules are parsed and validated
• Simulation runs deterministically
• JSON output is emitted

---

Project Layout (Current)

regsim/
  cli.py
  commands/
    simulate.py
  core/
    evaluator.py
    fields.py
    simulation.py
    validators.py
  schemas/

The CLI stays thin, while core rule evaluation lives under regsim/core/.

---

Rule Versioning & Regulatory Drift

RegSim-IN v1 supports explicit rule versioning:

• Rules declare:

  • rule_version
  • effective_from
  • source_reference

• No rule updates happen implicitly

• Simulations are always tied to a known regulatory snapshot

This ensures:

• Reproducibility
• Reviewability
• Trust

---

CI/CD Usage Example

regsim-in simulate --rules rules/ --input payload.json || exit 1

A failing rule causes a non-zero exit code.

---

Roadmap (Explicit, Not Promised)

Planned future directions include:

• Regulatory failure memory & correlation
• Safer rule authoring workflows
• Async system modeling hooks
• Community-contributed rule sets

These are not part of v1.

---

Philosophy

• Simulation over certification
• Explicit rules over implicit assumptions
• Deterministic behavior over magic
• Memory over repetition

---

Disclaimer

RegSim-IN is a developer simulation tool. It does not guarantee legal or regulatory compliance.

Always consult qualified professionals for real-world compliance decisions

---

Exact Commands

Run against JSON payload

regsim-in simulate --rules rules/ --input payload.json

Run against Python service

regsim-in simulate --rules rules/ --input src/

CI usage

regsim-in simulate --rules rules/ --input src/ --snapshot-date 2024-04-01