govengine 0.1.1

Carrier-agnostic governed execution services that consume SCLite contract lifecycle artifacts for policy-gated security automation.

GovEngine

GovEngine is a carrier-agnostic governed-execution core for policy-gated security automation.

It consumes SCLite as its contract lifecycle layer and provides reusable services around action validation, policy decisions, execution-contract shaping, execution-ticket checks, scope handling, command-shape normalization, and dry-run result assembly.

Project owner: Krzysztof Probola.

Why it exists

AI-assisted security workflows need a hard boundary between:

1. what an agent or caller wants;
2. what policy allows;
3. what execution shape was prepared;
4. what was approved;
5. what was dry-run or executed;
6. what evidence can be reviewed.

SCLite defines the auditable contract artifacts for that lifecycle. GovEngine is the reusable Python service layer that consumes those contracts and helps a host runtime enforce them without relying on prompt text alone.

Dependency direction

Ravenclaw -> GovEngine -> SCLite

• SCLite owns schema-backed lifecycle artifacts and validation.
• GovEngine owns reusable governed-execution helpers that consume SCLite artifacts.
• Ravenclaw remains the reference runtime/control plane and concrete integration host.

GovEngine is not Ravenclaw, Logdash, an LLM agent loop, a scanner, or a protocol adapter.

What GovEngine includes now

• action schema, validation, and compiler helpers;
• capability recipe and tool-registry helpers;
• semantic-loss classification helpers;
• policy core and policy-gateway helpers;
• execution-contract shaping/redaction helpers;
• approved-spec and execution-ticket validation helpers;
• command-shape and scope helpers;
• dry-run result assembly helpers;
• explicit SCLite integration seams;
• focused standalone pytest coverage and GitHub Actions CI.

What it intentionally does not include yet

• live subprocess execution backend;
• raw artifact storage/writes;
• Logdash UI/API routes;
• OpenClaw, MCP, A2A, or other protocol adapters;
• LLM provider integrations;
• Ravenclaw-specific personas, workspace state, or campaign UX;
• production-readiness claims.

Current status

GovEngine is pre-alpha extraction work. The package is importable and tested, and Ravenclaw has a migration branch that consumes it from this repository. The current public surface is intended for review and boundary hardening before any live execution backend is moved.

Installation

Once published to PyPI:

python -m pip install govengine

GovEngine depends on the PyPI distribution sclite-core while preserving the Python import package sclite.

For local development:

python -m venv .venv
. .venv/bin/activate
python -m pip install -e '.[dev]'
python -m pytest -q

Minimal smoke example

from govengine.action_compiler import compile_action_spec
from govengine.execution.runner import legacy_action_spec_dry_run_result

compiled = compile_action_spec({
    "action_type": "single_probe",
    "capability": "http_probe",
    "tool": "curl",
    "args": ["https://example.com"],
})

receipt = legacy_action_spec_dry_run_result(
    compiled_action=compiled,
    planned_commands=[["curl", "https://example.com"]],
)
assert receipt["status"] == "dry-run"

Documentation

• PUBLIC_STATUS.md — current maturity and non-claims.
• CHANGELOG.md — notable public changes.
• CONTRIBUTING.md — contribution and boundary rules.
• SECURITY.md — security reporting and package safety boundaries.
• PUBLISHING.md — publishing/PyPI readiness checklist.
• docs/ARCHITECTURE.md — package shape and dependency boundaries.
• docs/SCLITE_INTEGRATION.md — how GovEngine consumes SCLite.
• docs/API_BOUNDARY.md — owned vs excluded surfaces.
• docs/VALIDATION.md — local checks and non-claims.
• docs/ROADMAP.md — staged extraction roadmap.

Safety boundary

GovEngine should preserve deterministic governance over prompt-only behavior. Any future execution backend must be introduced behind explicit interfaces and tests, with Ravenclaw retaining the concrete runtime adapter until reviewed.