SoftTech AI Registry
Docusign · Docusign IAM

Iris Ready-to-Deploy Agents

Ready-to-deploy agents are the Docusign-provided, pre-built Iris agents for repeatable multi-step agreement tasks such as intake triage, playbook review and redlining, obligation tracking and relationship or renewal briefs. Users invoke them by chatting with the Iris assistant or add them as discrete steps in Workflow Builder. They are distinct from the Iris AI assistant itself, from customer-created custom agents built in Agent Studio, from Worksheets and from the Docusign MCP Server.

Recorded characteristics

Function
Docusign documents ready-to-deploy agents as pre-built agents, powered by the Docusign Iris AI engine, designed to execute complex multi-step agreement tasks out of the box, without a team designing an agent from scratch. Documented invocation paths: users can invoke them by chatting with the Iris assistant, or plug them directly into Workflow Builder as discrete steps. Developers can supply workflow inputs, connect the result to a downstream process, and orchestrate the surrounding experience with Workflow Builder, the Workflow Builder API or the Docusign MCP Server. Documented task areas for these agents: retrieving agreement information, monitoring status and obligations, applying legal playbooks, and guiding users through standard agreement processes; Docusign also lists intake triage, playbook review and obligation tracking as the workflows these agents are built for. Named ready-to-deploy agents on the current Docusign Agents page: - Easy intake and triage: automates agreement triage based on pre-defined risk profiles and exceptional terms, to accelerate approver review cycles. - Smart redlining: compares agreements against company standards and previously accepted terms and performs the redlining. - Relationship intelligence: assembles a comprehensive brief from signed agreements and deal history for quarterly business reviews, vendor reviews and negotiations. At the agents level (covering both ready-to-deploy and custom agents), Docusign states that agents can check agreements against company standards, suggest edits, automatically request the right approvals, monitor contracts in the background, flag risks, track obligations and trigger next steps without manual follow-up. Individual agents differ: no single ready-to-deploy agent is documented performing all of these operations, and this record does not attribute the full list to every agent.
Data access
Positively established: agreement context held by Docusign Iris, described as terms, obligations, amendments, negotiation history and renewal provisions; the repository of signed agreements; deal history; templates; approvals; and business logic. The named agents are documented reading agreements for triage against pre-defined risk profiles and exceptional terms, comparing agreements against company standards and previously accepted terms, and assembling briefs from signed agreements and deal history. Scope control: Docusign states that agents are limited to the agreement data the user can access, and that permissions configured in Docusign Agreement Manager and other Docusign IAM services determine which agreements and agreement sets that user can retrieve. Not attributed here: organization-specific playbooks, policies and connected third-party records such as Salesforce or HubSpot data through Extension Apps, which Docusign documents in the custom-agent and Agent Studio context rather than as ready-to-deploy agent access. Worksheets portfolio extraction across thousands of agreements is a separate Agreement Manager capability. Data reachable through the Docusign MCP Server in third-party AI applications is not attributed to these agents.
Actions
Can take actions
External actions
Unknown
Human confirmation
Not established
Permission basis
Not established
Administrative control
Documented controls and governance statements applying to these agents: - Permissions configured in Docusign Agreement Manager and other Docusign IAM services determine which agreements and agreement sets a user, and therefore an agent acting for that user, can retrieve. - Docusign states, for the agentic release as a whole, that every action is auditable and every workflow is governed, and positions policies, legal playbooks and compliance standards as the foundation of the agent trust layer. - Availability is gated by plan, data-center region and language, which functions as an account-level control over whether the agents are present at all. - Using agents as steps in Workflow Builder is gated to IAM Professional plans and above; Workflow Builder provides the rule-based routing, approval and conditional-logic scaffolding around an agent step. - Docusign advises builders designing agentic workflows to decide what data an agent can read or write, which actions it can perform, which actions require human confirmation, what data can be sent to external systems or models, and which actions and decisions need to be logged. This is stated as design guidance for builders, not as a documented product control specific to ready-to-deploy agents. Not established: a documented administrative surface for enabling, disabling, scoping or restricting individual ready-to-deploy agents; documented role restrictions on who may invoke them; and documented usage limits. The human-in-the-loop approvals and transparent audit trails described under Delegate with built-in oversight appear in the custom-agent section of the Docusign Agents page and are not imported here.
Default state
Not established
Availability
Verified against current Docusign first-party material, September 2026. Docusign states that the AI assistant, ready-to-deploy agents and custom agents in Agent Studio are generally available as of July 2026 for customers hosted in North America data centers with IAM Professional, IAM Enterprise, IAM for Sales or IAM Platform plans, English only. Using agents as steps in Workflow Builder is supported on IAM Professional plans and above, initially for accounts in North America data centers in English. Worksheets in Agreement Manager is stated as generally available as of August 2026 under the same conditions and is a separate capability. Docusign adds that availability varies by capability, plan, region, language and launch phase, and advises checking the latest product documentation before implementation. Earlier Momentum 2026 material describing early access in the U.S. from May 2026 with U.S. rollout starting in July is recorded as launch-stage history, not as current availability. The Docusign MCP Server is separately described as global open beta and is not part of this record.
Licensing
Included with the IAM plans listed above; Docusign publishes no separate price for ready-to-deploy agents in the material reviewed, and the Agents page directs prospects to Contact Sales. No pricing, consumption metering or seat model is inferred.
External model or provider
Not established
Limitations and uncertainty
ESTABLISHED: ready-to-deploy agents are Docusign-provided pre-built agents, distinct from the Iris AI assistant (the conversational entry point), from custom agents created in Agent Studio (customer-built and grounded in organizational policies and playbooks), from Worksheets (portfolio extraction in Agreement Manager), from Workflow Builder (the orchestration system) and from the Docusign MCP Server (bringing Docusign actions into third-party AI applications). Both invocation paths, Iris assistant chat and Workflow Builder steps, are documented. The three named ready-to-deploy agents and their described behaviour are documented on the current Docusign Agents page. ACTION CLASSIFICATION BASIS: action_capability is recorded as can_take_actions on the strength of specific documented operations rather than on the word agent. The Smart redlining agent is documented as doing the redlining for you, which produces changes to agreement content; the Easy intake and triage agent is documented as automating agreement triage against pre-defined risk profiles; and Docusign states that agents automatically request the right approvals and trigger next steps without manual follow-up, and that agents do not just find data, they take action. The precise persisted artefacts (for example whether a redline is stored as a document revision or a proposed change awaiting acceptance) are not specified in the material reviewed. WORKFLOW BUILDER BOUNDARY: an agent used as a Workflow Builder step returns a result that the workflow can use to route an agreement, request approval, escalate an exception or trigger another downstream action. Those downstream operations are performed by Workflow Builder under its configuration, not by the agent, and are deliberately not attributed to the agent here. NOT ESTABLISHED: human confirmation. Docusign describes agents requesting approvals and states that implementations should keep people in control of consequential decisions, but requesting an approval is not the same as requiring approval before the agent itself acts, and no documented per-action confirmation gate specific to ready-to-deploy agents was found; human_confirmation is therefore not_established rather than conditional. NOT ESTABLISHED: permission basis. Docusign establishes the read scope (agents are limited to the agreement data the user can access, governed by Agreement Manager and IAM permissions) but does not document the principal or authority under which an agent performs write or workflow operations, so permission_basis is not_established rather than user_permissions. NOT ESTABLISHED: external action. No consequential operation outside Docusign performed by a ready-to-deploy agent is positively documented. External authority is not inherited from the Docusign MCP Server, from Extension Apps, or from Salesforce, Slack, HubSpot or Agentforce integrations, and referencing third-party data is not a write; external_action_capability is unknown. NOT ESTABLISHED: operational default. Plan entitlement, general availability, presence in Iris and the ability to add an agent to a workflow are not treated as an operational default, so default_state is not_established. NOT ESTABLISHED: model or provider. Docusign attributes these agents to its proprietary Iris AI engine but does not identify the underlying model or provider used by ready-to-deploy agents. Frontier models named in MCP and integration material apply to those separate paths. CONFIGURATION AND VARIATION: which agents are available and what they do varies by plan, region, language and launch stage; individual ready-to-deploy agents have different functions; and behaviour inside a workflow depends on how the surrounding workflow is configured.

Evidence