Skip to content

Security: IndrasNet108/fsm-governance-engine-lib

Security

SECURITY.md

Security Policy

Scope

FSM Governance Engine is a validation-only library for deterministic verification of integrity-critical processes.

The project:

  • does not execute actions,
  • does not automate decisions,
  • does not enforce outcomes,
  • does not manage credentials, keys, or secrets,
  • does not perform network or filesystem operations.

Its sole responsibility is to validate process state transitions against formally defined rules and invariants.


Threat Model

The primary threat model addressed by this project is process integrity failure, including:

  • invalid or undocumented state transitions,
  • hidden or implicit process states,
  • out-of-order execution of governance or compliance steps,
  • lack of reproducible audit evidence,
  • human error in integrity-critical workflows.

The project does not aim to mitigate:

  • runtime compromise of the host system,
  • cryptographic attacks,
  • network-level attacks,
  • side-channel attacks,
  • denial-of-service scenarios.

Security Properties

The following properties are explicitly provided:

  • Deterministic behavior
    Identical inputs always produce identical validation outcomes.

  • Validation-only semantics
    The engine validates transitions but never performs side effects.

  • Explicit transition rules
    All allowed transitions are formally defined and exhaustively enumerated.

  • Invariant enforcement
    Declared invariants are checked as first-class validation constraints.

  • Fail-closed behavior
    Invalid transitions are rejected and surfaced as explicit errors.

  • Reproducibility
    Validation results can be reproduced using the same inputs and definitions.

  • Auditability
    Validation steps can produce structured, append-only audit records.


Out of Scope

The following are explicitly out of scope:

  • authorization or access control,
  • identity management,
  • key handling or cryptography,
  • policy enforcement,
  • workflow orchestration,
  • execution of business logic,
  • automated remediation.

Downstream systems are responsible for enforcing decisions and actions.


Secure Usage Guidelines

Consumers of this library are expected to:

  • treat validation failures as hard errors,
  • ensure FSM definitions are schema-validated before use,
  • persist audit artifacts in immutable or append-only storage where required,
  • apply independent authorization and execution controls outside this library,
  • review FSM definitions and invariants as part of change management.

Reporting Vulnerabilities

Security issues can be reported by email to:

info@indrasnet.ee

Please include:

  • a clear description of the issue,
  • steps to reproduce (if applicable),
  • affected versions or commit references.

We aim to acknowledge reports within a reasonable timeframe and coordinate disclosure where appropriate.


Security Updates

Security-relevant changes will be documented in the project changelog. No automatic update or notification mechanism is provided.


Disclaimer

This project provides validation primitives only. It is not a complete security solution and must be integrated into a broader security architecture.

There aren't any published security advisories