posture

Runtime-agnostic Python library for CCM (Continuous Control Monitoring) data collection. The entire contract: credentials in, DataFrame out. Runs unchanged in Docker, Airflow, Databricks — the library never knows or cares where it executes.

See docs/ARCHITECTURE.md for the design behind this library — the collect/parse split, locked design decisions, manifest schema, and per-collector implementation notes.

Installation

pip install posture

posture loads a .env file from the current directory (or a parent) automatically on import — no code changes needed. Variables already set in the environment always take precedence over .env values.

# .env
CROWDSTRIKE_CLIENT_ID=xxx
CROWDSTRIKE_CLIENT_SECRET=xxx
OKTA_DOMAIN=https://your-org.okta.com
OKTA_TOKEN=xxx
WORKSPACEONE_CLIENT_ID=xxx
WORKSPACEONE_CLIENT_SECRET=xxx
WORKSPACEONE_API_SERVER=asXXX.awmdm.com
WORKSPACEONE_TOKEN_URL=https://na.uemauth.workspaceone.com/connect/token  # optional, see below
UPGUARD_API_KEY=xxx
UPGUARD_BASE_URL=https://au.cyber-risk.upguard.com/api/public  # optional, see below
JAMF_URL=https://your-org.jamfcloud.com
JAMF_CLIENT_ID=xxx
JAMF_CLIENT_SECRET=xxx
INTUNE_TENANT_ID=xxx
INTUNE_CLIENT_ID=xxx
INTUNE_CLIENT_SECRET=xxx
MDE_TENANT_ID=xxx
MDE_CLIENT_ID=xxx
MDE_CLIENT_SECRET=xxx
AZURE_TENANT_ID=xxx
AZURE_CLIENT_ID=xxx
AZURE_CLIENT_SECRET=xxx
KNOWBE4_TOKEN=xxx
KNOWBE4_REGION=us  # optional, see below
TENABLEIO_ACCESS_KEY=xxx
TENABLEIO_SECRET_KEY=xxx
SALESFORCE_USERNAME=xxx
SALESFORCE_PASSWORD=xxx
SALESFORCE_TOKEN=xxx
SALESFORCE_DOMAIN=test  # optional, see below
SALESFORCE_SCHEMA_FILE=/path/to/salesforce.json  # optional, see below
QUALYS_USERNAME=xxx
QUALYS_PASSWORD=xxx
QUALYS_BASE_URL=https://qualysapi.qualys.com  # platform URL, varies by subscription
WIZ_CLIENT_ID=xxx
WIZ_CLIENT_SECRET=xxx
WIZ_API_ENDPOINT=https://api.us1.app.wiz.io/graphql
WIZ_TOKEN_URL=https://auth.app.wiz.io/oauth/token  # optional, see below
SAILPOINT_BASE_URL=https://your-tenant.api.identitynow.com
SAILPOINT_CLIENT_ID=xxx
SAILPOINT_CLIENT_SECRET=xxx
SNYK_TOKEN=xxx
SNYK_ENDPOINT=https://api.snyk.io  # optional, see below
DNSIMPLE_TOKEN=xxx
DNSIMPLE_ENDPOINT=https://api.dnsimple.com/v2/  # optional, see below
PHRIENDLY_PHISHING_CLIENT_ID=xxx
PHRIENDLY_PHISHING_CLIENT_SECRET=xxx
SERVICENOW_INSTANCE=acme
SERVICENOW_AUTH_TYPE=oauth2  # optional, defaults to oauth2 — see below
SERVICENOW_CLIENT_ID=xxx
SERVICENOW_CLIENT_SECRET=xxx
SERVICENOW_USERNAME=xxx
SERVICENOW_PASSWORD=xxx
SERVICENOW_SCHEMA_FILE=/path/to/servicenow.json  # optional, see below

Usage

from posture import CCM

ccm = CCM("crowdstrike")                          # creds from CROWDSTRIKE_* env vars
ccm = CCM("crowdstrike", {"client_id": "xxx"})    # partial override, rest from env

df = ccm.collect("hosts")                          # always a complete pandas DataFrame
ccm.flush_cache()                                  # the only cache invalidation

collect() always returns a complete pandas.DataFrame for the requested resource, or raises — there is no such thing as a partial snapshot in this library.

Discovering what's available

from posture import catalog

catalog()
# {
#   "crowdstrike": {
#     "required_config": {"client_id": "CROWDSTRIKE_CLIENT_ID", "client_secret": "CROWDSTRIKE_CLIENT_SECRET"},
#     "resources": {
#       "hosts": {"derived_from": None, "columns": ["client_id", "device_id", ...]},
#       "vulnerability_remediations": {"derived_from": "vulnerabilities", "columns": [...]},
#       ...
#     },
#   },
#   "knowbe4": {...},
#   ...
# }

catalog() never instantiates a collector, never touches the network, and needs no credentials — it reads sources, required config (as constructor key → env var), and resources (including which are derived, and their declared columns) straight off the registered Collector classes. It only reports required config — optional knobs (e.g. region, base_url) aren't tracked as data, so check a source's section below for those.

Example: export Crowdstrike hosts to local JSON

import json
from pathlib import Path

from posture import CCM

# CROWDSTRIKE_CLIENT_ID / CROWDSTRIKE_CLIENT_SECRET must be set in the environment
ccm = CCM("crowdstrike")
df = ccm.collect("hosts")

output_dir = Path("output")
output_dir.mkdir(exist_ok=True)

output_path = output_dir / "hosts.json"
output_path.write_text(df.to_json(orient="records", date_format="iso", indent=2))

print(f"Wrote {len(df)} hosts to {output_path}")

Supported sources

Crowdstrike configuration

Okta configuration

Workspace ONE configuration

UpGuard configuration

Tune vendor_risks with collect("vendor_risks", max_workers=8), or pass min_severity to filter server-side (see docs/ARCHITECTURE.md for why this one fans out per vendor).

Jamf configuration

computers_inventory_detail fetches one computer at a time by id, so expect one request per device on top of the initial listing call. See docs/ARCHITECTURE.md for which fields are ported.

Intune, MDE, and Azure Entra configuration

All three authenticate via Azure AD client-credentials against the tenant's OAuth2 endpoint (shared internal helper, not vendor SDKs).

Every collect() is a full snapshot — none of the three support incremental sync. mde's machine_vulnerabilities page size is overridable via a page_size kwarg. azure_entra's signins takes an optional days kwarg (default 180) that narrows the server-side $filter on createdDateTime — still a full point-in-time pull, not a checkpoint. See docs/ARCHITECTURE.md for endpoint-level detail on all three.

KnowBe4 configuration

pst_recipients (per-recipient phishing test results — delivered/opened/clicked/ reported timestamps) reads PST ids from psts internally unless a pst_ids kwarg is given; concurrency defaults to 10 workers, overridable via a max_workers kwarg.

Salesforce configuration

Requires the optional simple_salesforce dependency — install with pip install "posture[salesforce]". Auth is username + password + security token (no connected app / client id-secret needed).

Resources aren't hand-written per endpoint: salesforce.json declares one entry per Salesforce object as a flat {field_name: type} map, and both the SOQL query and the manifest are generated from that file. Add an object by editing the JSON (or pointing schema_file at your own), not by changing collector code.

ServiceNow configuration

Raw REST against the Table API — no vendor SDK. Supports two auth modes, chosen by auth_type (defaults to "oauth2" if unset):

oauth2 mode needs client_id/client_secret/username/password (ServiceNow's resource-owner password grant); basic mode needs only username/password.

ccm = CCM("servicenow")                              # oauth2 (default)
ccm = CCM("servicenow", {"auth_type": "basic"})       # basic auth instead

Resources aren't hand-written per endpoint: servicenow.json declares one entry per table as a flat {field_name: type} map, and the manifest (including the sysparm_fields list) is generated from that file. Add a table by editing the JSON (or pointing schema_file at your own), not by changing collector code. Filter results with the vendor's own query syntax via a sysparm_query kwarg, e.g. ccm.collect("cmdb_ci", sysparm_query="active=true").

Tenable.io configuration

Requires the optional pytenable dependency — install with pip install "posture[tenableio]".

Tenable.sc configuration

Requires the optional pytenable dependency — install with pip install "posture[tenablesc]".

vulnerabilities takes optional filters / tool kwargs (defaults: exclude informational severity, last seen in 30 days; vulndetails tool). hosts and asset_ips are scoped to a named Tenable.sc asset list via an asset_name kwarg (default "Non Crowdstrike Assets"). See docs/ARCHITECTURE.md for why asset_ips isn't a derived_from of assets.

Qualys configuration

Auth is HTTP Basic.

vulnerabilities here is Qualys' KnowledgeBase (the QID catalogue — severity, CVSS, CVE), not a per-host finding; vulnerability_detections is the per-host one. See docs/ARCHITECTURE.md for how both are fetched.

Wiz configuration

Direct cursor-paginated GraphQL queries (no report-export flow). See docs/ARCHITECTURE.md for a caveat on the GraphQL field paths used here.

SailPoint configuration

Targets Identity Security Cloud (ISC, the cloud SaaS product formerly known as IdentityNow) — not IdentityIQ. OAuth2 client-credentials against <base_url>/oauth/token, then offset/limit-paginated REST API v3 calls.

AppOmni configuration

Static bearer token auth (no OAuth flow). policies and posture_policies hit the same /policy/ endpoint with different default filters (reference policies vs. monitored-service-config policies). See docs/ARCHITECTURE.md for a caveat on the manifest field paths used here.

Snyk configuration

Static token auth (Authorization: token ...). members, projects, and issues have no "all orgs" endpoint, so each fans out per organisation id across a thread pool — org ids are read from organizations internally unless an org_ids kwarg is given; concurrency defaults to 8 workers, overridable via a max_workers kwarg. See docs/ARCHITECTURE.md for a caveat on the manifest field paths used here.

Cloudflare configuration

Static API token auth (Authorization: Bearer ...), global API base URL (no tenant subdomain). dns_records and cdn_protected_domains have no "all zones" endpoint, so each fans out per zone id across a thread pool — zone ids are read from zones internally unless a zone_ids kwarg is given; concurrency defaults to 8 workers, overridable via a max_workers kwarg. cdn_protected_domains hits the same /zones/{zone_id}/dns_records endpoint as dns_records with proxied=true passed server-side, returning only the records actually routed through Cloudflare's CDN. See docs/ARCHITECTURE.md for a caveat on the manifest field paths used here.

DNSimple configuration

Static bearer token auth. Every v2 endpoint is scoped under an account id, so _authenticate calls DNSimple's whoami once to discover it before the first request. See docs/ARCHITECTURE.md for a caveat on the manifest field paths used here.

PhriendlyPhishing configuration

OAuth2 client-credentials against a dedicated auth host (auth.api.phriendlyphishing.com), separate from the API host. clicks defaults its server-side start_time/end_time range to the trailing 366 days (plus one day forward); pass start_time/end_time kwargs (YYYY-MM-DD) to override. See docs/ARCHITECTURE.md for a caveat on the manifest field paths used here.

Vanta configuration

OAuth2 client-credentials against Vanta's global token host (https://api.vanta.com/oauth/token) — no tenant subdomain or regional discovery. Every resource is a real top-level paginated endpoint (cursor-based pageSize/pageCursor), no fan-out. See docs/ARCHITECTURE.md for a caveat on the manifest field paths used here.

Development

pip install -e ".[dev]"
pytest
ruff check src tests
black src tests