llmhq-releaseops 0.1.0

Release engineering infrastructure for AI behavior - bundle, promote, evaluate, and replay agent behavior artifacts

llmhq-releaseops

Release engineering for AI agent behavior — bundle prompts, policies, and model configs into versioned artifacts, then promote them through gated environments.

Why This Exists

AI agents ship behavior through prompts, policies, and model configurations — not deterministic code. When something breaks in production, there's no git blame for "why did the agent start approving refunds it shouldn't?" ReleaseOps brings standard release engineering (bundle, promote, rollback, observe) to these behavior artifacts, so you always know what's running, what changed, and why.

Install

pip install llmhq-releaseops

Extra	Install	Adds
eval	pip install llmhq-releaseops[eval]	LLM-as-judge (OpenAI, Anthropic)
langsmith	pip install llmhq-releaseops[langsmith]	LangSmith trace queries
dev	pip install llmhq-releaseops[dev]	pytest, black, mypy

Quickstart

1. Initialize your project:

releaseops init

Creates .releaseops/ with environments (dev, staging, prod), bundle storage, and eval directories.

2. Create a bundle from your prompts and model config:

releaseops bundle create support-agent \
  --artifact system=onboarding:v1.2.0 \
  --model claude-sonnet-4-5 --provider anthropic

3. Promote through environments:

releaseops promote promote support-agent 1.0.0 dev
releaseops promote promote support-agent 1.0.0 staging
releaseops promote promote support-agent 1.0.0 prod

4. Load at runtime in your agent code:

from llmhq_releaseops.runtime import RuntimeLoader

loader = RuntimeLoader()
bundle, metadata = loader.load_bundle("support-agent@prod")
# bundle.model_config, bundle.prompts, bundle.policies — all resolved
# metadata is automatically injected into OpenTelemetry spans

5. Compare versions when something changes:

releaseops analytics compare support-agent@1.0.0 support-agent@1.1.0

Key Concepts

• Bundle — immutable, content-addressed manifest of prompts + policies + model config (SHA-256 verified)
• Environment — named deployment target (dev/staging/prod) with a pinned bundle version
• Promotion — moving a bundle through environments with optional quality gates (eval, approval, soak)
• Telemetry — automatic injection of bundle metadata into OpenTelemetry spans for production observability
• Attribution — trace agent behavior back to specific prompt lines and policy rules that caused it

Documentation

• IMPLEMENTATION.md — full CLI reference, Python SDK guide, data model reference, architecture details

License

MIT