Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Replo Public 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%2Freplo.app%2Freplo" | 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%2Freplo.app%2Freplo" | 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 Replo Public API.
Start a Replo agent session on a project from a prompt and answer its pending permission interactions
Publish a site, promoting only selected routes such as /about or a blog post when you pass promoteRoutes
Connect a custom domain to a site and poll DNS verification until it activates
Create and update products with a title, slug, and externalCheckoutUrl, and set per-variant stock with trackInventory and available
GET STARTED
List orders filtered by financialStatus such as paid or refunded, or by customerEmail
Run read-only SELECT queries against storefront analytics, including session_attribution revenue
Schedule recurring agent tasks with a prompt, scheduledAt, and rrule
Patterns agents use Replo Public API for, with concrete tasks.
★ Your agent drives Replo's site agent
A coding or ops agent connected through Jentic opens a Replo agent session with a projectId and a prompt, sends follow-up messages, polls session status, and resolves pending permission interactions with approve or reject. The page edits happen inside Replo; your agent orchestrates and checks the result.
Call POST /agent/sessions with projectId and prompt set to 'Add a testimonials section below the hero', then GET /agent/sessions/{sessionId} until it finishes
Controlled publishing
After edits are reviewed, a site is published with promoteRoutes listing only the routes that should go live, up to 200 of them, while other page drafts stay unpublished. Passing an empty array publishes only site-wide changes.
Call POST /sites/{siteId}/publish with promoteRoutes set to ["/pricing"]
Product and stock updates
Merchants create Replo products with a title, description, slug, and externalCheckoutUrl, then set per-variant inventory with trackInventory and available counts, keeping storefront pages in step with stock.
Call POST /products with projectId and title set to 'Classic Hoodie', then PATCH /products/{productId}/inventory with trackInventory true and available set to 40 for each variant
Storefront revenue analysis
Growth teams run read-only SQL through the analytics query operation, for example against session_attribution(since, until) to see attributed revenue by page path. Queries are limited to SELECT or WITH statements, 60 seconds of execution, and 100000 result rows, with project scoping enforced server-side.
Call POST /analytics/query with projectId set to the project and query set to a SELECT summing attributedRevenueUsd by path from session_attribution('2026-10-01', '2026-10-31')
40 endpoints — the replo public api builds and runs replo storefront sites: it starts replo agent sessions that edit a project from a natural-language prompt, publishes sites with optional promoteroutes, connects and verifies custom domains, and creates and updates products with per-variant inventory.
METHOD
PATH
DESCRIPTION
/agent/sessions
Create an agent session
/agent/sessions/{sessionId}
Get agent session status
/sites/{siteId}/publish
Publish a site
/custom-domains
Connect a custom domain to a site
/products
Create a product
/products/{productId}/inventory
Set per-variant inventory
/orders
List orders
/analytics/query
Run a read-only analytics query
/agent/sessions
Create an agent session
/agent/sessions/{sessionId}
Get agent session status
/sites/{siteId}/publish
Publish a site
/custom-domains
Connect a custom domain to a site
/products
Create a product
/products/{productId}/inventory
Set per-variant inventory
/orders
List orders
/analytics/query
Run a read-only analytics query
What agents get from Jentic-routed access to this vendor.
Setup
Wiring Replo by hand means sending its bearer token plus the Replo-Api-Version header on each call, polling agent sessions and domain verification, and following cursor pagination. Through Jentic you install once, import Replo from the API Directory, store the token once, and your agent calls it.
Permission scoping
Replo puts site, session, product, and task ids in the URL path (/sites/{siteId}/publish, /products/{productId}), so a rule can pin your agent to one site or one product. You choose the operations it may call, so destructive ones like custom domain deletion, task deletion, or minting API keys are not included unless you add them.
Credential isolation
Your Replo token 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 'publish a Replo site' or 'prompt the Replo agent', and Jentic returns the matching Replo operation with its input schema so the agent sends the right ids and fields.
Alternatives and complements available in the Jentic catalogue.
Specific to using Replo Public API through Jentic.
What authentication does the Replo Public API use?
Per its OpenAPI spec, the Replo Public API authenticates with an HTTP bearer token in the Authorization header. Requests also send a Replo-Api-Version header, set to 2026-10-05 in this spec. Minting a new key through POST /api-keys requires the api_keys.write scope when called with an API key. Through Jentic, the token is stored encrypted by your own Jentic One instance and injected at execution time.
Can I publish only some pages of a Replo site?
Yes. POST /sites/{siteId}/publish accepts promoteRoutes, an array of up to 200 routes such as / or /blog/my-post. Only those routes' unpublished edits go live; omit it to publish everything with changes, or pass an empty array to publish only site-wide changes.
How do I prompt the Replo agent through the API?
Call POST /agent/sessions with a projectId and a prompt. Send follow-ups with POST /agent/sessions/{sessionId}/messages, check progress with GET /agent/sessions/{sessionId}, resolve permission interactions with decision approve or reject, and stop a run with the abort operation.
What are the limits on Replo analytics queries?
Per the spec, POST /analytics/query accepts SELECT or WITH queries only, with a maximum execution time of 60 seconds and at most 100000 result rows. Project scoping is enforced server-side, so queries must not add project filters.
What are the rate limits for the Replo Public API?
The OpenAPI spec does not specify request rate limits. Check https://replo.app for current limits.
How do I connect a custom domain to Replo through Jentic?
Search Jentic for 'connect a custom domain to a Replo site'. Jentic returns POST /custom-domains with its input schema; your agent sends siteId and domain, then calls the verify operation until DNS checks pass. To run it on your own infrastructure, install Jentic One from its GitHub repo.
Is there a Replo MCP server?
You don't need an MCP server to give your agent Replo. Jentic connects it directly from the API Directory: import it, store your token once, and your agent finds the session, publishing, and product operations by intent instead of loading every tool definition into its context.
Can I limit what my agent is allowed to do with the Replo Public API?
Yes. A content agent could be allowed POST /agent/sessions, GET /agent/sessions/{sessionId}, and POST /sites/{siteId}/publish for one siteId, so it can edit and ship that site but cannot touch products, orders, custom domains, or API keys. Leave DELETE /custom-domains/{customDomainId} and POST /api-keys out of its rule.
For Agents
Prompt Replo's site agent to edit a project, publish sites, connect custom domains, manage products and inventory, read orders, and query storefront analytics with SQL.
Use for: Ask the Replo agent to add a testimonials section to our landing page, Publish the Replo site but only push the /pricing route, Connect shop.example.com to our Replo site, List all refunded Replo orders this week
Not supported: Does not process payments, ship orders, or edit Shopify store settings. Use for building, publishing, and managing Replo sites, products, and storefront analytics only.
The Replo Public API builds and runs Replo storefront sites: it starts Replo agent sessions that edit a project from a natural-language prompt, publishes sites with optional promoteRoutes, connects and verifies custom domains, and creates and updates products with per-variant inventory. It also lists orders filtered by financialStatus or customerEmail, runs read-only SQL analytics queries such as session_attribution revenue, imports assets, stores project files, and schedules recurring agent tasks with an rrule. Every request carries a Replo-Api-Version header set to 2026-10-05.
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>