ambertraceai 0.5.0

Python SDK for the Ambertrace neurosymbolic AI platform API

AmbertraceAI Python SDK

Python client for the Ambertrace neurosymbolic AI platform API.

Install

pip install ambertraceai

Authentication

The SDK authenticates with an Ambertrace API key (prefix at_...). Create one from the dashboard at app.ambertrace.ai → Settings → API Keys, then pass it to the client:

from ambertraceai import AmbertraceAPI

api = AmbertraceAPI(base_url="https://app.ambertrace.ai", api_key="at_...")

Keep the key out of source control — read it from an environment variable in real code:

import os
api = AmbertraceAPI(base_url="https://app.ambertrace.ai", api_key=os.environ["AMBERTRACE_API_KEY"])

See Agent Keys for the user- vs. platform-scoped key model.

Quick Start

from ambertraceai import AmbertraceAPI

api = AmbertraceAPI(
    base_url="https://app.ambertrace.ai",
    api_key="at_...",
)

# Create a domain
domain = api.domains.create(
    name="Legal Contracts",
    description="Contract analysis for risk and compliance",
)

# Upload data
dataset = api.datasets.upload(
    domain_id=domain["id"],
    file_path="contracts.csv",
)

# Build a platform (async — returns the platform and a build job)
result = api.platforms.create(
    domain_id=domain["id"],
    dataset_id=dataset["id"],
)
platform_id = result["platform"]["id"]
build_job_id = result["build_job"]["id"]

# Wait for the build to finish
job = api.wait_for_job(build_job_id, timeout=600)

# Query the platform
answer = api.platforms.query(
    platform_id=platform_id,
    query="What are the highest-risk clauses?",
)
print(answer["answer"])
print(answer["explanation"])

Resources

Resource	Methods
api.domains	list, create, get, update, delete, build_ontology, eval_config, set_eval_config, delete_eval_config, suggest_eval_config, list_templates, create_template, update_template, delete_template, feedback_stats
api.datasets	list, get, upload, fetch, quality, clean, preview, delete
api.platforms	list, create, get, delete, update, status, query, suggest_rules, list_suggestions, approve_suggestion, reject_suggestion, graph, list_rules, create_rule, update_rule, delete_rule, capture_drift_baseline, check_drift
api.predictions	predict, list_configs, create_config, delete_config, train, list_predictions
api.connectors	list, test
api.usage	get
api.jobs	get
api.api_keys	list, create, revoke

Connectors

Connectors pull data from external providers. List what's available, optionally test a config, then ingest it as a dataset linked to a domain:

api.connectors.list()   # discover connectors + their required config fields

# Stocks/ETFs and crypto are keyless:
api.datasets.fetch(domain_id=1, connector_type="yahoo",
                   config={"symbols": ["AAPL", "SPY"], "range": "2y"})
api.datasets.fetch(domain_id=1, connector_type="coinbase",
                   config={"product_ids": ["BTC-USD", "ETH-USD"]})

# FRED needs your own free key (https://fred.stlouisfed.org):
api.datasets.fetch(domain_id=1, connector_type="fred",
                   config={"api_key": "<your FRED key>",
                           "series_ids": ["GS10", "FEDFUNDS"], "frequency": "monthly"})

# Generic REST/CSV — bring your own auth via headers:
api.datasets.fetch(domain_id=1, connector_type="rest",
                   config={"url": "https://api.example.com/series",
                           "headers": {"Authorization": "Bearer ..."}})

Connector	Config	Key?
yahoo	symbols, interval, range	none
coinbase	product_ids, granularity	none
fred / fred_sentiment	series_ids, frequency, api_key	bring your own
rest	url, format, records_path, headers, params	bring your own (via headers)

Bring your own provider keys. Connectors that hit a credentialed provider require your own key, passed in config — Ambertrace never uses a shared key on your behalf.

Agent Keys

AI agents authenticate with user-scoped API keys that give full lifecycle access (domains, datasets, platforms, rules, predictions). A human creates the key from the dashboard; the agent can then create narrower platform-scoped keys for its integrations.

# Agent creates a platform-scoped key for a specific integration
platform_key = api.api_keys.create(
    scope="platform",
    platform_id=42,
    name="Slack Integration",
)

# List keys visible to this agent
keys = api.api_keys.list()

# Revoke a platform key the agent created
api.api_keys.revoke(platform_key["id"])

User-scoped keys cannot create other user-scoped keys (no self-replication). Chat, conversations, and billing remain human-only.

Verified Profile

Platforms can be built with the verified profile for proof-carrying queries:

result = api.platforms.create(
    domain_id=domain["id"],
    dataset_id=dataset["id"],
    verified_profile=True,
    verified_min_confidence=0.85,
    invariant_manifest=[
        {"name": "no_unconditional_delete", "kind": "forbid",
         "target": "permit_delete", "assumed_absent": ["permit_delete"]},
    ],
)

# Query — response includes proof certificate
answer = api.platforms.query(platform_id, query="What permissions does alice have?")
print(answer["proof_checked"])   # True
print(answer["proof_summary"])   # Human-readable certificate

# Drift monitoring
api.platforms.capture_drift_baseline(platform_id)
drift = api.platforms.check_drift(platform_id)
print(drift["drift_detected"])   # False

Rules CRUD

Manage symbolic rules on a platform:

rules = api.platforms.list_rules(platform_id)

rule = api.platforms.create_rule(
    platform_id,
    name="viewer_read_only",
    condition={"field": "role", "operator": "==", "value": "viewer"},
    action={"type": "derive", "value": "read_only"},
)

api.platforms.update_rule(platform_id, rule["id"], description="Updated desc")
api.platforms.delete_rule(platform_id, rule["id"])

Job Polling

Long-running operations (platform builds, data cleaning, training) return a job_id. Use wait_for_job to poll:

job = api.wait_for_job(job_id, timeout=300, poll_interval=5)
if job["status"] == "error":
    print(f"Failed: {job.get('error_message')}")

Error Handling

from ambertraceai import AmbertraceAPI, AmbertraceError

try:
    api.domains.get(999)
except AmbertraceError as e:
    print(e.status_code)  # 404
    print(e.code)         # "not_found"
    print(str(e))         # "Domain not found."

API Documentation

Full API reference: app.ambertrace.ai/openapi/redoc