OPEN-SOURCE CONTROL LAYER / PYTHON 3.10+

Govern AI agents
before they act.

Define what your AI agents can and cannot do before they act. Lightweight, framework-agnostic, and production-ready.

$ pip install hlinor-registry

View on GitHub Read Documentation

policy trace / 001 runtime ready

One reviewable contract. One decision before side effects.

50+

automated tests

3.10–3.13

Python CI matrix

CrewAI

& LangChain ready

A policy is a file.
A decision is a line.

Keep governance close to the code, readable in a pull request, and enforceable at runtime with zero friction.

examples/agent.yaml YAML / registry

id: web-research-agent
name: Web Research Agent
department: marketing
enforcement_mode: strict
skills:
  - web_search
  - summarize_text
policies:
  - max_10_searches_per_hour
  - require_budget_check
allowed_actions:
  - search
  - read_public_url
  - extract_keywords
blocked_actions:
  - login_to_website
  - call_premium_paid_api

terminal / zero-friction cli pre-dispatch

$ hlinor-registry check --bundle bundle.json --agent web-research-agent --action search

[ALLOWED]EXPLICITLY_ALLOWED

$ hlinor-registry explain --bundle bundle.json --agent web-research-agent --action call_premium_paid_api

[DENIED]ACTION_BLOCKLISTED_VIOLATED_POLICY_REQUIRE_BUDGET_CHECK

✗ Action is explicitly listed in blocked_actions
HOW TO FIX: Remove from blocked_actions and recompile.

$ hlinor-registry check ... --format jsonl
{"timestamp": "2026-07-25T...", "agent_id": "web-research-agent", "result": "deny", "decision_id": "a1b2c3..."}

Native wrappers available for LangChain and CrewAI.

DEVELOPER EXPERIENCE

Drop-in integration for your stack.

No need to rewrite your agent logic. Just wrap your existing tools or use our Python decorator for zero-boilerplate enforcement.

app/tools.py Python

from hlinor_registry.integrations.decorators import governed

@governed(
    agent_id="research-agent", 
    action="search_web", 
    bundle_path="bundle.json"
)
def search_web(query: str) -> str:
    # This function will simply not execute if the policy denies it.
    # A PolicyViolationError is raised with a clear reason_code.
    return requests.get(f"https://api.search.com?q={query}").text

Governance that answers the hard questions early.

01 / PRIVACY

Prevent PII leaks

Block external email, record mutation, or other exfiltration paths for agents handling sensitive data.

02 / COST

Control API costs

Keep search counts, premium calls, budget checks, and rate limits visible beside the agent definition.

03 / EVIDENCE

Ensure compliance

Export runtime decisions in JSONL format for direct ingestion into Splunk or Datadog. Lint YAML policies statically to catch logical contradictions before deployment.

A clear path from intent to action.

Registry files become a shared contract between the people who design agent capabilities and the systems that execute them.

Make your next agent easier to trust.

Start with a YAML contract, add a policy check before the first side effect, and invite your security team into the review.

Star the repository

OPEN SOURCE / EARLY ACCESS

HlinorAI works with teams designing internal agent platforms, compliance workflows, and production action boundaries.

Govern AI agents before they act.

A policy is a file. A decision is a line.