Rox

A documented integration set with write back: the documentation covers Salesforce integration and writeback, HubSpot, email, calendar, Slack, Microsoft Teams and data warehouse integrations (Snowflake, Databricks, BigQuery), and the Teams plan lists CRM sync and writeback, so agents act in outside systems with the customer's credentials (docs.rox.com index and rox.com/pricing).

Sourcedocs.rox.com/development/product/integrations/salesforce/configuring-salesforce-integration-+-writeback.mdread 2026-10-01

Multi step, multi agent work the buyer configures: teams build custom agents through documented creation paths, attach triggers and outputs, and run them beside the prebuilt research, outreach, deal and meeting agents, while the vendor's agent framework documents account aware swarms orchestrating research, outreach, opportunity management and proposal generation (docs.rox.com custom agents and agent framework pages).

Sourcedocs.rox.com/development/product/custom-agents.mdread 2026-10-01

The corpus is the customer's own and the ingestion is documented. A unified, governed knowledge graph consolidates structured sources including CRM, email and calendar data with unstructured ones including support tickets, job postings and news feeds, plus finance and product telemetry, and the graph is built natively inside the customer's own warehouse so entities, relationships and context across accounts, opportunities and interactions can be extracted reliably. Every agent action reasons over that graph rather than over generic context.

Sourcemedium.com/snowflake/rox-the-next-generation-revenue-operating-system-powered-by-snowflake-08b5c6a9a963read 2026-09-01

A person controls whether an agent runs and can judge what it produced: custom agents run only once a team switches the Active toggle on, after a test run, and outputs carry approve and reject controls. No documented step holds a CRM write or email for that approval before it commits, so the approve control is review of output rather than a gate on an action (docs.rox.com custom agents pages).

Sourcedocs.rox.com/development/product/custom-agents/activate-monitor-and-review-outputs.mdread 2026-10-01

The documentation's Governance section describes Permission Sets bundling resource, field and custom resource permissions assigned to users or through Pods, organization wide defaults for record and field access, and Access Provenance for tracing who can view a record; Enterprise organizations set up SAML SSO with their own identity provider through a self serve wizard. The security page states SOC 2 Type II compliance with annual independent audits, and the trust center lists a SOC 2 Type 2 report on request and CASA Tier 2, with ISO 27001 marked coming soon.

SourceRox docs governance and enterprise SSO pages, rox.com/security and trust.rox.com; docs.rox.com/development/product/governance/permission-sets.mdread 2026-09-30

Agent runs are on the record: each custom agent is monitored from its Agent page, a Runs page and the Home page, and its outputs (text, HTML, email, slides, files) can be reviewed, approved or rejected. The documentation does not say what a run record shows, inputs, steps or tool calls, so why an agent acted cannot be reconstructed; agentic monitoring tracks account signals, not agent runs (docs.rox.com custom agents and monitoring pages).

Sourcedocs.rox.com/development/product/custom-agents/activate-monitor-and-review-outputs.mdread 2026-10-01

Agent memory distinct from the knowledge graph is not documented: nothing describes what an account agent retains from one cycle to the next, no scope or lifetime for agent memory is stated, and the architecture is described as agents reasoning over the graph rather than writing to a memory they later read. The persistent understanding each account agent builds is that governed knowledge graph, the customer's own corpus.

Sourceaws.amazon.com/blogs/machine-learning/rox-accelerates-sales-productivity-with-ai-agents-powered-by-amazon-bedrockread 2026-09-01

Warehouse-native execution is the defining architectural property and the vendor states it plainly in its own marketplace listing: all customer data stays inside the customer's data warehouse.

A technical post co-authored by Rox engineers explains the mechanism, using Snowflake zero-copy data sharing to gain governed access to warehouse data without persisting raw datasets outside customer control, with the Rox knowledge graph grounded natively in the customer's Snowflake instance.

That places the customer's data in their own environment and region by construction, a stronger residency position than naming regions and well beyond shared multi-tenant software as a service. The Rox application layer itself runs on AWS.

Sourceaws.amazon.com/marketplace/pp/prodview-hzr2usullmhoaread 2026-09-01

A named set of prebuilt agents, each with its own documentation page and job: the Research, Outreach, Deal, Meet, Autofill and Apps agents, beside custom agents a team builds (docs.rox.com index and custom agents section).

Sourcedocs.rox.com/llms.txtread 2026-10-01

Custom agents wake on documented triggers without a person: on a schedule (every weekday at 8 AM), before or after a meeting, on transcript completion, on opportunity activity, on a webhook call from outside Rox, and on a Clever Column cell being generated; tracked Clever Columns also re-run on a schedule and notify on change (docs.rox.com agent triggers and agentic monitoring pages).

Sourcedocs.rox.com/development/product/custom-agents/agent-triggers.mdread 2026-10-01

The vendor discloses the model layer as Amazon Bedrock, which its agents run on, but Bedrock is the infrastructure the platform is built on rather than a choice surfaced to the buyer, in the same way a cloud provider on a subprocessor list is infrastructure rather than a second model. No model selector, no per-agent choice, no bring-your-own-key and no routing policy is documented in any vendor-authored material, and no provider or model is offered for the customer to pick. Which models its agents use on Bedrock is not published.

Sourceaws.amazon.com/blogs/machine-learning/rox-accelerates-sales-productivity-with-ai-agents-powered-by-amazon-bedrockread 2026-09-01

An outside system can start a Rox agent: custom agents take an On webhook call trigger for events that start outside Rox, an inward interface to the platform. No API reference, keys page, SDK or MCP server is listed in the documentation index, so the webhook trigger is the only documented way in (docs.rox.com agent triggers page and index).

Sourcedocs.rox.com/development/product/custom-agents/agent-triggers.mdread 2026-10-01

A customer runs a test of a custom agent before switching it on, and reviews outputs on the Runs page afterwards, which is inspection rather than a scored check. The Agent Evaluation System the documentation describes, with labeled datasets, accuracy metrics and LLM as a judge rubrics, is run by Rox's own developers on Rox's agents. No harness, scored test case or release gate a customer runs on its own agents is documented (docs.rox.com custom agents and agent evaluation system pages).

Sourcedocs.rox.com/development/product/custom-agents/activate-monitor-and-review-outputs.mdread 2026-10-01

No action on an interface the vendor does not control is documented. The vendor's own listing describes agents creating personalized emails and LinkedIn posts, which is generating content for a person to send or publish rather than the agent operating LinkedIn.

Everything else runs over programmatic routes: a knowledge graph built by warehouse sharing, connectors to CRM, support and finance systems, and delivery through web, Slack, macOS and iOS clients. No hosted or local browser, desktop session, remote computer control, scraping or third-party browser engine is documented in any vendor-authored material.

Sourcemarketplace.microsoft.com/en-us/product/rox-com.rox-enterprise-revenue-ai-agentsread 2026-09-01