Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Decionis Commerce Gate 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%2Fdecionis.com%2Fdecionis" | 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%2Fdecionis.com%2Fdecionis" | 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 Decionis Commerce Gate API.
Validate one bounded ERP transaction synchronously through the Commerce Gate before it runs
Record tenant-bound commerce action evaluations in Shadow Mode without blocking execution
Read signed Decision Dossiers and their verification proof packets for one tenant
Review Shadow Mode reports and aggregate decision outcomes
GET STARTED
Assess a bounded previous-90-day commerce history without executing any commerce action
Evaluate a labelled synthetic sample in a verified AgentSaaS trial
Patterns agents use Decionis Commerce Gate API for, with concrete tasks.
★ Guarding an AI Agent's Commerce Actions
An AI agent connected to Decionis through Jentic can submit a bounded ERP transaction to the guard and get a synchronous validate decision before it acts, so a risky or out-of-policy commerce action is caught before execution rather than after. Because the organization is bound to the credential server-side, the agent cannot evaluate actions for a tenant other than its own.
Submit a bounded ERP purchase transaction to the guard validate operation and proceed only if the decision allows it
Shadow-Mode Evaluation Before Enforcement
Teams can run Decionis in Shadow Mode to record how the gate would have decided on real commerce actions without blocking any of them. Each evaluation is captured tenant-bound, and the Shadow Mode reports aggregate outcomes so a team can measure the gate's behaviour before switching it to synchronous enforcement.
Record a commerce action evaluation in Shadow Mode and then read the aggregate Shadow Mode report for the tenant
Signed Decision Audit Trail
Every evaluated decision can be read back as a signed Decision Dossier with a verification proof packet, giving an immutable, tenant-bound record of what was decided and why. This supports audit and dispute review, since the proof packet lets a third party verify the dossier without trusting the caller.
Read a Decision Dossier by its id and fetch the matching proof packet to verify the recorded outcome
Licensed Historical Commerce Assessment
On the licensed AWS Commerce surface, Decionis assesses a bounded previous-90-day history from connected-store and synthetic sources and returns an immutable assessment bound to the current organization and AWS account. The assessment runs without executing any commerce action, so it reviews past activity rather than changing anything.
Run a bounded previous-90-day history assessment against a licensed source and read the resulting immutable assessment by id
9 endpoints — the decionis commerce gate api validates bounded erp and commerce transactions synchronously before they run, acting as a guard between an agent and the actions it wants to take.
METHOD
PATH
DESCRIPTION
/v1/guard/validate
Validate a bounded ERP transaction through the Commerce Gate
/v1/protocol/evaluate-decision
Record a commerce action evaluation in Shadow Mode
/v1/protocol/dossiers/{dossierId}
Read a signed Decision Dossier
/v1/protocol/dossiers/{dossierId}/proof-packet
Read the verification proof packet for a dossier
/v1/protocol/shadow-reports
Read Shadow Mode reports and aggregate outcomes
/aws/commerce/history/assessments
Assess a bounded previous-90-day history
/aws/commerce/history/sources
List licensed historical assessment sources
/v1/guard/validate
Validate a bounded ERP transaction through the Commerce Gate
/v1/protocol/evaluate-decision
Record a commerce action evaluation in Shadow Mode
/v1/protocol/dossiers/{dossierId}
Read a signed Decision Dossier
/v1/protocol/dossiers/{dossierId}/proof-packet
Read the verification proof packet for a dossier
/v1/protocol/shadow-reports
Read Shadow Mode reports and aggregate outcomes
/aws/commerce/history/assessments
Assess a bounded previous-90-day history
/aws/commerce/history/sources
List licensed historical assessment sources
What agents get from Jentic-routed access to this vendor.
Setup
Wiring Decionis by hand means handling three credential shapes (an X-Decionis-API-Key header, a Decionis bearer token, and an AWS Marketplace bearer key) whose organization is bound server-side. Through Jentic you install once, import Decionis from the API Directory, store the credential once, and your agent calls it.
Permission scoping
The Commerce Gate separates its operations: synchronous validation and Shadow Mode evaluation, dossier reads, and the licensed 90-day history assessment are distinct calls. A rule can let your agent validate transactions and read dossiers while the licensed assessment operations stay out of its reach. You choose the operations it may call, and the tenant is already fixed server-side to the credential.
Credential isolation
Your Decionis API key is stored once, encrypted, by your own Jentic One instance and injected at execution time. It never enters the agent's prompt, logs, or context.
Intent-based discovery
Agents search Jentic by intent such as 'validate a commerce transaction' or 'read a decision dossier', and Jentic returns the matching Decionis 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 Decionis Commerce Gate API through Jentic.
Is there a Decionis Commerce Gate MCP server?
Decionis is designed to be reached through a local Commerce Gate MCP adapter, but you don't have to stand one up to use it from your agent. Jentic connects the Commerce Gate directly from the API Directory: import it, store your credential once, and your agent calls the guard and protocol operations, discovered on demand so no extra tool definitions bloat its context.
Can I limit what my agent is allowed to do with the Decionis Commerce Gate API?
Yes. Write a rule that allows the guard validate and Shadow Mode evaluation operations and dossier reads but not the licensed 90-day history assessment, so the agent can check and record commerce decisions while the licensed assessment surface stays off, and every call it makes is logged.
What authentication does the Decionis Commerce Gate API use?
Per its OpenAPI spec, the API accepts a Decionis bearer token or an X-Decionis-API-Key header for the Commerce Gate and protocol operations, and a separate AWS Marketplace bearer key for the licensed Commerce surface. In every case the organization and license are bound server-side to the credential, so no caller-supplied org id is accepted. Through Jentic the credential is stored encrypted by your own Jentic One instance and injected only at call time.
Can I validate an ERP transaction before it runs with the Decionis Commerce Gate API?
Yes. The guard validate operation takes one bounded ERP transaction and returns a synchronous decision before the action is executed, and the evaluate-decision operation records the same kind of evaluation in Shadow Mode when you do not want to block. Each decision can be read back as a signed Decision Dossier with a proof packet.
What are the rate limits for the Decionis Commerce Gate API?
The OpenAPI spec does not specify request-rate limits; the licensed AWS Commerce surface instead bounds assessments to a previous-90-day window per the contract. Check decionis.com for current limits and licensing details.
How do I guard a commerce action with the Decionis Commerce Gate API through Jentic?
Call the guard validate operation with the bounded transaction, act only on an allowing decision, and read the Decision Dossier afterward for an audit record. In Jentic, search for 'validate a commerce transaction through the gate', store your credential once, and the agent runs the check inline. To run it on your own infrastructure, install Jentic One from its GitHub repo.
For Agents
Validate ERP and commerce transactions through a synchronous guard, record decisions in Shadow Mode, and read signed Decision Dossiers with proof packets. Covers licensed 90-day history assessment across 9 endpoints.
Use for: Validate an ERP transaction through the Commerce Gate before executing it, Record a commerce action evaluation in Shadow Mode, Read the signed Decision Dossier for a decision id, Get the verification proof packet for a dossier
Not supported: Does not process payments, place orders, or execute ERP transactions itself. Use for validating and auditing commerce decisions through the Commerce Gate only.
The Decionis Commerce Gate API validates bounded ERP and commerce transactions synchronously before they run, acting as a guard between an agent and the actions it wants to take. It records tenant-bound decision evaluations in Shadow Mode, issues signed Decision Dossiers with verification proof packets, and aggregates outcomes into Shadow Mode reports. On the licensed AWS Commerce surface it assesses a bounded previous-90-day history and evaluates labelled synthetic samples in a verified trial, without executing commerce actions itself.
This API is usable in Jentic One now. Its AI-readiness score against Jentic's framework shows where it stands today and where improvements would make it even easier for agents to use.
Base layer of spec validity and structural soundness.
Aggregated quality score from linter diagnostics, weighted by severity.
Percentage of `$ref` references that resolve successfully.
Checks whether the API description parses successfully and conforms to its declared specification (e.g., OpenAPI).
Structural correctness score based on schema issues using logarithmic dampening.
Clarity, completeness, and ingestion readiness for developers and tooling.
How richly the API is illustrated with examples.
Percentage of examples that conform to their schemas.
Percentage of operations with complete response definitions (success, client error, server error).
Health of API ingestion, bundling, and resolution within Jentic pipelines.
Semantic breadth, depth, and agent comprehension for AI systems.
Coverage of descriptions across API elements.
Coverage of RFC 9457 Problem Details for error responses.
Coverage, uniqueness, and casing consistency of operationIds for AI inference.
Coverage of summaries across operations/tags/info.
Functional utility, complexity comfort, and AI orchestration readiness.
Agent comfort level based on API operational and structural complexity.
Trust, risk posture, and security compliance.
Average quality of security schemes based on authentication method strength (weakest link for OAuth2).
Findability, semantic richness, and reasoning readiness.
Clarity and depth of descriptions across API elements.
Score it yourself
Every API in the directory is allowlisted, so you can re-score it with no key required.
npx @jentic/api-scorecard-cli score <openapi-url>