BIGHUB
The execution governance layer for autonomous AI agents.
Ask AI about BIGHUB
Powered by Claude Β· Grounded in docs
I know everything about BIGHUB. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
BIGHUB
Decision learning for AI agents.
BIGHUB is a decision layer that evaluates agent actions, returns structured recommendations, and learns from real outcomes over time.
Agent proposes action
β
BIGHUB evaluates in context
β
returns recommendation + confidence + rationale
β
agent / runtime decides what to do
β
real outcome is reported
β
future recommendations improve over time
Packages
| Package | Language | Install | Description |
|---|---|---|---|
| bighub | Python | pip install bighub | Core SDK β evaluate actions, receive recommendations, report outcomes, retrieve learned signals. |
| bighub-openai | Python | pip install bighub-openai | OpenAI Responses API adapter β evaluate tool calls, surface recommendations, learn from outcomes. |
| @bighub/bighub-mcp | TypeScript | npm install @bighub/bighub-mcp | MCP server β evaluate agent actions from any MCP client. |
| bighub-anthropic | Python | β | Anthropic adapter β coming soon. |
| bighub-openai (JS) | TypeScript | β | OpenAI adapter for Node.js β coming soon. |
How it works
| Static guardrails | BIGHUB |
|---|---|
| Fixed allow / block logic | Structured recommendation in context |
| Rules only | Constraints + precedents + calibration + outcomes |
| One-off judgment | Trajectory- and history-aware evaluation |
| Prediction only | Prediction compared with real outcomes |
| Static thresholds | Recommendation quality improves over time |
Quickstart (Python SDK)
pip install bighub
import os
from bighub import BighubClient
client = BighubClient(api_key=os.getenv("BIGHUB_API_KEY"))
# 1. Submit an action for evaluation
result = client.actions.submit(
action="update_price",
value=150.0,
domain="pricing",
actor="AI_AGENT_001",
)
# 2. Inspect the recommendation
print(result["recommendation"]) # proceed, proceed_with_caution, review_recommended, do_not_proceed
print(result["recommendation_confidence"]) # high, medium, low
print(result["risk_score"]) # 0.0 β 1.0
# 3. Act based on the recommendation
if result["recommendation"] in ("proceed", "proceed_with_caution"):
execute_action()
# 4. Report the real outcome
client.outcomes.report(
request_id=result["request_id"],
status="SUCCESS",
description="Price updated, no negative impact observed",
)
elif result["recommendation"] == "review_recommended":
request_human_review()
client.close()
Core loop: evaluate β recommend β act β report outcome β learn.
Structured recommendation
BIGHUB primarily returns:
recommendationβ what to do (proceed,proceed_with_caution,review_recommended,do_not_proceed)recommendation_confidenceβ how confident (high,medium,low)risk_scoreβ aggregated risk (0β1)enforcement_modeβ how the recommendation is applied (advisory,review,enforced)decision_intelligenceβ rationale, evidence status, trajectory health, alternatives
Legacy fields such as allowed, result, and reason may still appear for backward compatibility, but they are not the primary product surface.
Trajectory-aware evaluation
BIGHUB evaluates actions not only in isolation, but also in the context of what happened before. As outcomes accumulate, similar sequences and prior decisions improve future recommendations.
For costly and multi-step workflows, trajectory-aware signals mean the same action may be judged differently depending on what happened earlier in the sequence.
Free BETA
Current Free BETA limits:
- 3 agents
- 2,500 actions / month
- 30 days history
- 1 environment
Repository layout
βββ sdk/
β βββ python/
βββ adapters/
β βββ python/
β β βββ openai/
β β βββ anthropic/
β βββ js/
β βββ openai/
βββ servers/
β βββ mcp/
βββ examples/
Links
- bighub.io
- GitHub β bighub-io/bighub
- PyPI β bighub
- PyPI β bighub-openai
- npm β @bighub/bighub-mcp
