Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the IBM Operational Decision Manager API, or any other public or private API you need. You set the rules, the agent never sees your credentials, and every call is logged.
Two steps, two machines. Install the instance in a safe environment, then register your agent from wherever it runs.
Step 1: Jentic One Host machine
# On the machine that will host your Jentic One instance:
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fibm.com%2Fodm-api" | shStep 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL "https://jentic.com/install.sh?src=apis&api=%2Fapis%2Fibm.com%2Fodm-api" | sh
jentic register # connects your agent to your Jentic One instanceJentic One is in public beta. The setup above keeps your agent separate from the instance, which is what you want before using real credentials: an agent running as the same OS user as Jentic One can read its stored keys directly. Just evaluating? A single local install is fine to start. See the secure deployment guide for the tiers.
What an agent can do with IBM Operational Decision Manager API.
List and inspect rule applications via /v2/ruleapps and /v2/ruleapps/{ruleAppName}
Execute the latest version of a ruleset via /v2/ruleapps/{ruleAppName}/{rulesetName}/execute
Execute a specific ruleset version via /v2/ruleapps/{ruleAppName}/{rulesetVersion}/{rulesetName}/execute
GET STARTED
For Agents
Execute IBM ODM business rules and decision services - call rulesets and decision services with input payloads and get back rule-engine outputs.
Use for: I need to execute an ODM ruleset for a loan eligibility check, Get the result of a decision service for a pricing input, List all rule applications deployed on ODM, Execute a specific version of an ODM ruleset
Not supported: Does not handle ruleset authoring, deployment from Decision Center, or rule debugging - use for executing and inspecting deployed ODM rulesets and decision services only.
IBM Operational Decision Manager (ODM) is the rule engine and decision platform IBM ships for executing business rules at runtime, used in pricing, eligibility, fraud, and routing decisions. The REST API exposes both the Rule Execution Server (RES) for executing rulesets and the decision services surface for invoking decision models, so external systems and agents can call deterministic decisions without embedding the rule engine. It is suited to teams that want to reuse rule-based decisions across many channels.
List and execute decision services via /v1/decisionservices and /v1/decisionservices/{dsId}/execute
Inspect server status and diagnostics via /v2/status and /v2/diagnostics
Patterns agents use IBM Operational Decision Manager API for, with concrete tasks.
★ Eligibility and Pricing Decisions
Call an ODM ruleset to make a deterministic eligibility or pricing decision - for example, deciding whether an applicant qualifies for a loan or what discount tier applies. POST the input payload to /v2/ruleapps/{ruleAppName}/{rulesetName}/execute and ODM returns the rule-engine output, keeping the decision logic in a maintained ruleset instead of scattered application code.
POST a JSON body with applicant income and loan amount to /v2/ruleapps/loans/eligibility/execute and parse the decision and explanation from the response.
Versioned Ruleset Execution
Pin an integration to a specific ruleset version while a new version is being tested in parallel. The /v2/ruleapps/{ruleAppName}/{rulesetVersion}/{rulesetName}/execute endpoint takes the version explicitly, which lets release managers run A/B comparisons or roll forward and back without reconfiguring callers.
POST the same payload to /v2/ruleapps/loans/1.2/eligibility/execute and /v2/ruleapps/loans/1.3/eligibility/execute and compare the two outputs.
Decision Service Invocation
Invoke a higher-level ODM decision service that may compose multiple rulesets and data lookups. The /v1/decisionservices/{dsId}/execute endpoint is the entry point for consumers that should not know the internal rule structure, which makes it the right surface to expose to less-privileged callers.
POST the customer profile to /v1/decisionservices/fraudCheck/execute and use the returned risk score to decide whether to clear the transaction.
AI Agent Decision Caller
An AI agent uses Jentic to call ODM rulesets when a decision must be deterministic and auditable rather than LLM-generated - pricing tiers, eligibility cuts, or compliance rules. The agent searches by intent, loads the execute schema, and posts the input, receiving a rule-engine answer with an explanation. Typical agent integration is under one hour.
Search Jentic for 'execute an ODM ruleset', load the /v2/ruleapps/{ruleAppName}/{rulesetName}/execute schema, and run the eligibility ruleset for a candidate applicant.
12 endpoints — ibm operational decision manager (odm) is the rule engine and decision platform ibm ships for executing business rules at runtime, used in pricing, eligibility, fraud, and routing decisions.
METHOD
PATH
DESCRIPTION
/v2/ruleapps
List rule applications
/v2/ruleapps/{ruleAppName}
Get a rule application
/v2/ruleapps/{ruleAppName}/{rulesetName}/execute
Execute the latest version of a ruleset
/v2/ruleapps/{ruleAppName}/{rulesetVersion}/{rulesetName}/execute
Execute a specific ruleset version
/v1/decisionservices
List decision services
/v1/decisionservices/{dsId}/execute
Execute a decision service
/v2/status
Get server status
/v2/diagnostics
Get server diagnostics
/v2/ruleapps
List rule applications
/v2/ruleapps/{ruleAppName}
Get a rule application
/v2/ruleapps/{ruleAppName}/{rulesetName}/execute
Execute the latest version of a ruleset
/v2/ruleapps/{ruleAppName}/{rulesetVersion}/{rulesetName}/execute
Execute a specific ruleset version
/v1/decisionservices
List decision services
What agents get from Jentic-routed access to this vendor.
Setup
Wiring the IBM Operational Decision Manager API by hand means configuring basic auth against your Decision Service host and port and mapping its ruleset execution routes yourself. Through Jentic you install once, import the IBM Operational Decision Manager API from the API Directory, store the credentials once, and your agent calls it.
Permission scoping
ODM puts the rule app and decision service ids in the URL path (/ruleapps/{ruleAppName}, /decisionservices/{dsId}), so a rule can pin your agent to inspecting and executing one deployed rule app. You choose the operations it may call, so executing a decision service is not included unless you add it.
Credential isolation
Your ODM credentials are stored once, encrypted, by your own Jentic One instance and injected at execution time. They never enter the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'execute a deployed ruleset' or 'list available rule apps', and Jentic returns the matching operation with its input schema so the agent calls the right endpoint without browsing the reference docs.
Alternatives and complements available in the Jentic catalogue.
Specific to using IBM Operational Decision Manager API through Jentic.
What authentication does the IBM ODM API use?
The ODM REST API uses HTTP basic authentication. Through Jentic, those credentials are stored in your Jentic One instance and the agent calls ODM via a scoped Jentic credential, so the basic auth header is never visible in the agent's context.
Can I execute a specific version of a ruleset?
Yes. POST the input payload to /v2/ruleapps/{ruleAppName}/{rulesetVersion}/{rulesetName}/execute with the version path segment. This is the right call when you need to pin an integration to a known ruleset version while a new version is being tested.
What is the difference between rulesets and decision services?
Rulesets under /v2/ruleapps/{ruleAppName} are the unit deployed by the Rule Execution Server, while decision services under /v1/decisionservices are higher-level entry points that may compose multiple rulesets and data sources. Pick the surface that matches the abstraction the caller should know about.
How do I call an ODM ruleset through Jentic?
Search Jentic for 'execute an ODM ruleset', load the /v2/ruleapps/{ruleAppName}/{rulesetName}/execute schema, and execute with the input payload. Install the SDK with pip install jentic and use the async search, load, execute pattern.
Can I check the health of the ODM server through this API?
Yes. GET /v2/status returns the live status of the Rule Execution Server and /v2/diagnostics returns deeper diagnostics. These are the right endpoints to wire into a readiness or health check ahead of routing rule traffic.
What are the rate limits for the IBM ODM API?
ODM does not enforce a fixed REST rate limit; throughput depends on the Rule Execution Server configuration and the rule complexity. Treat 429 and 503 responses as backpressure, retry with exponential backoff, and prefer batched payloads where the ruleset supports them.
Can I limit what my agent is allowed to do with the IBM ODM API?
Yes. Because you self-host Jentic One, your own rules decide which ODM operations and credentials the agent may use, and ODM puts the rule app and decision service ids in the URL path (/ruleapps/{ruleAppName}, /decisionservices/{dsId}), so you can pin the agent to inspecting and executing a single deployed rule app. You choose the exact operations it can call, so executing a decision service under /v1/decisionservices/{dsId}/execute is not available to the agent unless you add it. This lets you allow, for example, only /v2/ruleapps/{ruleAppName}/{rulesetName}/execute for one rule app while keeping every other endpoint out of reach.
/v1/decisionservices/{dsId}/execute
Execute a decision service
/v2/status
Get server status
/v2/diagnostics
Get server diagnostics