Jentic is an AI infrastructure company building the agentic knowledge layer for APIs. Founded in late 2024 and backed by $4.5M in pre-seed funding, Jentic enables enterprises to confidently manage, scale, and govern AI agent initiatives in a unified platform built on open standards. The platform provides secure execution, managed authentication, unified permissions, and observability for AI agents accessing 1,500+ public APIs and 2,000+ agent-ready workflows. Jentic's bet is that AI agents need a standards-based control plane built on OpenAPI plus Arazzo workflows plus MCP, rather than ad-hoc tool wiring inside every agent framework. The platform ships the Jentic Public APIs catalog, the Arazzo Engine (workflow runner), the Standard Agent (a composable ReWOO-style reasoning agent), the Jentic Mini self-hosted execution layer, a hosted Remote MCP server at api.jentic.com/mcp, and the Jentic API AI-Readiness Framework (JAIRF) for scoring API readiness for agents.
Jentic publishes 39 API contracts indexed on the APIs.io network, including Authentication API, Execution API, Search API, and 36 more. Tagged areas include AI Agents, Arazzo, OpenAPI, MCP, and Workflows.
The Jentic catalog on APIs.io includes 1 JSON-LD context and 2 Spectral governance rulesets.
Jentic’s developer surface includes authentication, engineering blog, documentation, pricing, signup flow, developer console, YouTube channel, and 28 more developer resources.
Regulatory Posture applies to this provider. Its tags matched the
Horizontal (data, software, accessibility, platform) regime, so
Regulatory Posture carries 15 points of the composite.
If this regime is wrong for your business, say so on your
provider repo — the
applicability map is public and we will correct it.
Create-or-Update Ergonomics applies to this provider. This API accepts writes, so it
carries 10 points of the composite. It is scored from the published contracts
themselves: whether a caller can create-or-update in one call, whether the write accepts a key the caller already
holds, and whether the response says which branch ran. Without that, every write needs a search-and-branch in
front of it, and the first time that check is skipped a duplicate record is created.
Scored against the observed mean rather than raw — a provider at the catalog average is unchanged by this facet,
not penalised by it.
The six quality facets above are damped to 75 points between them,
because the conditional facet above carries the other
25. That is why each facet's contribution is shown against a damped
maximum: raising a quality facet moves the composite by 75% of its nominal
weight, not 100%. The full arithmetic is at apis.io/rating/.
Show this score on your site.
It updates itself when the score changes. Paste this where you link to your API docs:
<a href="https://apis.io/providers/jentic/"><img src="https://apis.io/badge/jentic/card.svg" alt="Jentic on APIs.io — Kin Score by API Evangelist"></a>
Other styles, a README pill and a shields.io endpoint are on the badge page.
Raise your score.
Agent readiness — checks not met, largest first (exact, from this profile):
Idempotency+9 pts — OpenAPI mutating operations declare an idempotency key — a header, a request-body property, or a named de-duplication token whose own description says so (verified); else common[].type matches /idempoten/i (documented)
A2A Agent Card+8 pts — a2a/-a2a.yml manifest, graded on conformance.grade against A2A 1.0.0
Delegated User Identity+6 pts — a served discovery document listing authorization_code in grant_types_supported (served), else an oauth2 authorizationCode flow or openIdConnect declared in an OpenAPI (documented)
Typed Event Surface+6 pts — AsyncAPI document, OR webhooks/callbacks objects in the as-published OpenAPI, OR common[].type includes Webhooks
Protected Resource Metadata+5 pts — a served /.well-known/oauth-protected-resource carrying `resource` (documented), and naming its authorization_servers (verified)
4 more unmet checks in the Agent Readiness table above.
Kin Score — facets with the most composite points still available (weight × headroom):
The Kin Score Market layer reads a provider inside its market:
its position, and how deep its integrations go. Depth is only comparable within a position. Provisional and unweighted:
not part of the Kin Score or agent readiness.
Registered MCP server and declares compatibility are report only — not weighted.
Undeclared means we found nothing we can read, so there is no number: we never infer integrations.
Standards this provider implements or carries. De facto ones are profiled and measured by the API Commons; formal ones come from the standards directory and are linked by the provider's own tags. The evidence column says how each claim was established — not that it was made.
The Jentic Remote MCP Server is the hosted Model Context Protocol endpoint that exposes the Jentic catalog to MCP-capable clients (Claude Desktop, ChatGPT, Cursor, Windsurf, VS ...
Open-source Python engine that executes Arazzo workflow specifications alongside OpenAPI operation definitions. Powers the Jentic platform's workflow execution and is also insta...
Open-source composable reasoning agent framework that plans, acts, and recovers from failures, preconfigured with Jentic tools and a ReWOO reasoning loop. Primary entry point is...
Free, open-source, self-hosted API execution layer that sits between an agent and the outside world. The agent says what it wants to do and Jentic Mini handles the how: finding ...
Technical specification for evaluating how interpretable, operable, and trustworthy an API is for AI systems and autonomous agents. Defines the signals, dimensions, scoring mode...
Actor-agnostic, multi-item access-request surface (Core / Access bounded context). When an agent's brokered call is rejected by a permission rule — or the agent needs access it ...
Unified actor directory — a lightweight read-only view across all actor types (users, agents, service accounts). Returns ID-to-name mappings for UI cache hydration so dashboards...
Agent actors — autonomous principals that call the Broker. Covers the agent lifecycle (list, read, approve / deny, enable / disable, archive) and the toolkits bound to each agent.
Sub-resource of **APIs** (Registry bounded context). Enumerate the operations exposed by a registered API — either the live (`current_revision_id`) view or a specific revision. ...
Sub-resource of **APIs** (Registry bounded context). Download the underlying OpenAPI document for a registered API — either the live (`current_revision_id`) view or a specific r...
The `Api` aggregate root — what APIs are registered for this deployment, addressable by the loose `(vendor, name, version)` tuple. Every successful import lands a new immutable ...
Append-only administrative audit trail — who did what, when. Distinct from `Executions` (brokered upstream calls); this records control-plane mutations.
The public API **catalog** — a registry-side browse/preview/import surface over the Jentic public-APIs GitHub manifest. Powers the Discover experience: list importable APIs (`GE...
Runtime, DB-backed platform configuration. Lets an operator set, read, and list configuration that previously required hand-editing backend YAML and restarting the server — star...
Part of the **Core / Access** bounded context — the authoritative record of "who can call what, with which secret". A `Credential` is the auth secret + header configuration the ...
Curated, severity-tagged event stream surfaced to dashboards and operators. Each event is either **informational** (no action required, e.g. `import.completed`) or **actionable*...
Append-only audit log of every Broker call (sync or async) — timing, status, trace IDs, and the upstream API reference (`vendor:name:version`). Bodies are not stored. Records ar...
Identity introspection for the calling principal (human, agent, or service account) — `GET /me` returns the resolved subject, scopes, and permissions behind the presented token.
Resolve a single registered operation to its full structural detail — parameters, response schema, security scheme, server URL. Heavier than `Lookup` (which is the Broker hot pa...
In-flight async-job tracker for both Registry import jobs and Broker async-execution jobs — one uniform `Job` shape keyed by `kind`. Lifecycle only — type-specific result payloa...
MCP (Model Context Protocol) transport reporting. The `jentic mcp` stdio server terminates all MCP protocol traffic locally; the control plane only sees plain HTTP. This tag cov...
Dashboard-oriented aggregation endpoints — execution volume, success/failure ratios, and top operations over a bounded time window. Results are cached in-process (TTL ~120 s) so...
Free-form annotations attached to registry resources — an `Api`, an `ApiRevision`, a registered `operation_id`, an `ExecutionRecord`, or a `Credential`. Used to capture human or...
OAuth 2.0 / OIDC endpoints exposed by the platform authorization server — the authorize, token, introspection, revocation, and assertion-mint endpoints plus the redirect callback.
Admin-managed registry of third-party OAuth clients (confidential, secret-bearing). Registered clients integrate with Jentic One via the standard Authorization Code + PKCE flow....
OpenAPI Overlay 1.0 (`https://spec.openapis.org/overlay/v1.0.0`) documents that correct an imported spec without rewriting it. Overlays capture wrong required fields, missing se...
Part of the **Admin / Audit** bounded context — the catalogue of grantable permission strings. Permissions are namespaced `resource:action` strings (`users:write`, `capabilities...
Operational endpoints for service health and platform tooling. Not part of any bounded context — surfaced here so orchestrators and load balancers have a stable probe target.
Sub-resource of **Toolkits** (Core / Access bounded context). A toolkit doesn't own credentials directly — it **binds** to credentials that already exist in `/credentials`. The ...
Sub-resource of **Toolkits** (Core / Access bounded context). A toolkit can hold many API keys at once — different agents, machines, or environments typically each get their own...
Part of the **Core / Access** bounded context — a `Toolkit` is a scoped bundle of credentials and permissions issued to an agent or service. This tag covers **toolkit lifecycle*...
Part of the **Admin / Audit** bounded context — the human roster of the organisation. One organisation, many users; admin surfaces here cover create / list / read / update / del...
aid: jentic
url: https://raw.githubusercontent.com/api-evangelist/jentic/refs/heads/main/apis.yml
name: Jentic
description: Jentic is an AI infrastructure company building the agentic knowledge layer for APIs. Founded in late 2024 and
backed by $4.5M in pre-seed funding, Jentic enables enterprises to confidently manage, scale, and govern AI agent initiatives
in a unified platform built on open standards. The platform provides secure execution, managed authentication, unified permissions,
and observability for AI agents accessing 1,500+ public APIs and 2,000+ agent-ready workflows. Jentic's bet is that AI agents
need a standards-based control plane built on OpenAPI plus Arazzo workflows plus MCP, rather than ad-hoc tool wiring inside
every agent framework. The platform ships the Jentic Public APIs catalog, the Arazzo Engine (workflow runner), the Standard
Agent (a composable ReWOO-style reasoning agent), the Jentic Mini self-hosted execution layer, a hosted Remote MCP server
at api.jentic.com/mcp, and the Jentic API AI-Readiness Framework (JAIRF) for scoring API readiness for agents.
deliveryModel:
model: saas
license_evidence:
not_product:
- jentic/api-ai-readiness-framework
- jentic/arazzo-engine
- jentic/jentic-mini
- jentic/standard-agent
open_source: unknown
commercial: true
callable_host: true
label: Hosted service · you call their endpoint
confidence: low
source:
- openapi
- pricing
- repository-not-product
generated: '2026-09-25'
method: derived
accessModel:
pricing: enterprise
onboarding: unknown
trial: false
try_now: false
public: false
label: Enterprise
confidence: medium
source:
- plans
- authentication
- security
generated: '2026-11-06'
method: derived
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/icons/jentic.png
type: Index
access: 3rd-Party
position: Consuming
created: '2026-01-02'
modified: '2026-05-19'
x-docs-platform: mkdocs
x-docs-platform-confidence: medium
x-docs-platform-checked: 2026-10-03
tags:
- AI Agents
- Arazzo
- OpenAPI
- MCP
- Workflows
- Integration
- Agent Runtime
- Standard Agent
- Just In Time Tooling
- Credential Vault
- Agent Governance
- Observability
- API AI Readiness
tags_raw:
- AI Agents
- Arazzo
- OpenAPI
- MCP
- Workflows
- Integrations
- Agent Runtime
- Standard Agent
- Just In Time Tooling
- Credential Vault
- Agent Governance
- Observability
- API AI Readiness
segments:
- AI Agents
- Workflows
- Integrations
- MCP
- Agent Infrastructure
specificationVersion: '0.23'
apis:
- aid: jentic:jentic-mcp
name: Jentic Remote MCP Server
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
description: The Jentic Remote MCP Server is the hosted Model Context Protocol endpoint that exposes the Jentic catalog
to MCP-capable clients (Claude Desktop, ChatGPT, Cursor, Windsurf, VS Code). It speaks the same search / load / execute
pattern as the Jentic API, but packaged as MCP tools. Supports OAuth 2.0 with PKCE and dynamic client registration, with
an API key fallback for Cursor and Windsurf.
tags:
- MCP
- AI Agents
- OAuth
- Remote MCP
- Authentication
tags_raw:
- MCP
- AI Agents
- OAuth
- Remote MCP
- Authentication
properties:
- url: https://docs.jentic.com/guides/mcp/
type: Documentation
- url: https://docs.jentic.com/guides/mcp/remote-mcp/
type: GettingStarted
- url: https://api.jentic.com/mcp
type: BaseURL
properties_authentication:
- name: OAuth 2.0
scheme: oauth2
notes: Recommended. PKCE supported. Automatic token refresh. Dynamic client registration. Used by Claude Desktop, ChatGPT,
Cursor, Windsurf.
- name: API Key
scheme: apiKey
notes: Fallback for Cursor and Windsurf. Uses an agent-scoped Jentic API key (JENTIC_AGENT_API_KEY).
properties_features:
- name: search_apis Tool
description: Lists matching APIs and workflows for a natural language query.
- name: load_execution_info Tool
description: Returns the schema and auth requirements for a chosen operation.
- name: execute Tool
description: Executes an operation or workflow with managed credentials.
- name: list_credentials Resource
description: Lists the agent's available upstream credentials.
- aid: jentic:arazzo-engine
name: Arazzo Engine
humanURL: https://github.com/jentic/arazzo-engine
description: Open-source Python engine that executes Arazzo workflow specifications alongside OpenAPI operation definitions.
Powers the Jentic platform's workflow execution and is also installable standalone as arazzo-runner for direct use in
agents and automation. Handles authentication resolution (API key, OAuth 2.0, basic, bearer), dynamic server URL resolution,
parameter chaining between steps, and blob storage for large binary payloads.
tags:
- Arazzo
- OpenAPI
- Workflows
- Open Source
tags_raw:
- Arazzo
- OpenAPI
- Workflows
- Open Source
- Open-Source
properties:
- url: https://github.com/jentic/arazzo-engine
type: SourceCode
- url: https://docs.jentic.com/getting-started/arazzo-runner/
type: Documentation
- url: https://pypi.org/project/arazzo-runner/
type: PythonPackage
- aid: jentic:standard-agent
name: Jentic Standard Agent
humanURL: https://github.com/jentic/standard-agent
description: Open-source composable reasoning agent framework that plans, acts, and recovers from failures, preconfigured
with Jentic tools and a ReWOO reasoning loop. Primary entry point is the ReWOOAgent Python class. Bring-your-own LLM (Anthropic,
OpenAI, Google). Demonstrates how to build an agent that uses Jentic's just-in-time tooling without committing to a specific
agent framework.
tags:
- Agent Framework
- ReWOO
- Open Source
- Python
tags_raw:
- Agent Framework
- ReWOO
- Open Source
- Python
- Open-Source
properties:
- url: https://github.com/jentic/standard-agent
type: SourceCode
- url: https://docs.jentic.com/getting-started/standard-agent/
type: Documentation
- aid: jentic:jentic-mini
name: Jentic Mini
humanURL: https://github.com/jentic/jentic-mini
description: 'Free, open-source, self-hosted API execution layer that sits between an agent and the outside world. The agent
says what it wants to do and Jentic Mini handles the how: finding the right API, injecting credentials at runtime, and
brokering the request. Intended for teams that want Jentic''s execution model without sending traffic through the hosted
Jentic API.'
tags:
- Self-Hosted
- Open Source
- Agent Runtime
tags_raw:
- Self Hosted
- Open Source
- Agent Runtime
- Open-Source
properties:
- url: https://github.com/jentic/jentic-mini
type: SourceCode
- aid: jentic:jairf
name: Jentic API AI-Readiness Framework (JAIRF)
humanURL: https://github.com/jentic/api-ai-readiness-framework
description: Technical specification for evaluating how interpretable, operable, and trustworthy an API is for AI systems
and autonomous agents. Defines the signals, dimensions, scoring model, and normalization rules used by the Jentic API
Scorecard product. Apache 2.0.
tags:
- Specification
- API AI Readiness
- Open Source
tags_raw:
- Specification
- API AI Readiness
- Open Source
- Open-Source
properties:
- url: https://github.com/jentic/api-ai-readiness-framework
type: SourceCode
- aid: jentic:jentic-authentication-api
name: Jentic Authentication API
description: Registration and API key management for accessing the Jentic platform.
humanURL: https://docs.jentic.com/reference/public-apis/
baseURL: https://api.jentic.com/api/v1
tags:
- Authentication
properties:
- type: OpenAPI
url: openapi/jentic-authentication-api-openapi.yml
- type: Website
url: https://jentic.com/
- type: Documentation
url: https://docs.jentic.com/
- type: GettingStarted
url: https://docs.jentic.com/getting-started/quickstart/
- type: APIReference
url: https://docs.jentic.com/reference/public-apis/
- type: Signup
url: https://app.jentic.com/sign-up
- type: Console
url: https://app.jentic.com/
- type: BaseURL
url: https://api.jentic.com/api/v1
- type: SpectralRules
url: rules/jentic-rules.yml
- type: JSONSchema
url: json-schema/jentic-operation-schema.json
- type: JSONSchema
url: json-schema/jentic-agent-schema.json
- type: JSONSchema
url: json-schema/jentic-workflow-schema.json
- type: JSONStructure
url: json-structure/jentic-operation-structure.json
- type: Examples
url: examples/jentic-search-example.json
- type: Examples
url: examples/jentic-load-example.json
- type: Examples
url: examples/jentic-execute-example.json
- type: Examples
url: examples/jentic-register-example.json
- aid: jentic:jentic-execution-api
name: Jentic Execution API
description: Load execution details and execute API operations or Arazzo workflows with managed authentication and credential
injection.
humanURL: https://docs.jentic.com/reference/public-apis/
baseURL: https://api.jentic.com/api/v1
tags:
- Execution
properties:
- type: OpenAPI
url: openapi/jentic-execution-api-openapi.yml
- type: Website
url: https://jentic.com/
- type: Documentation
url: https://docs.jentic.com/
- type: GettingStarted
url: https://docs.jentic.com/getting-started/quickstart/
- type: APIReference
url: https://docs.jentic.com/reference/public-apis/
- type: Signup
url: https://app.jentic.com/sign-up
- type: Console
url: https://app.jentic.com/
- type: BaseURL
url: https://api.jentic.com/api/v1
- type: SpectralRules
url: rules/jentic-rules.yml
- type: JSONSchema
url: json-schema/jentic-operation-schema.json
- type: JSONSchema
url: json-schema/jentic-agent-schema.json
- type: JSONSchema
url: json-schema/jentic-workflow-schema.json
- type: JSONStructure
url: json-structure/jentic-operation-structure.json
- type: Examples
url: examples/jentic-search-example.json
- type: Examples
url: examples/jentic-load-example.json
- type: Examples
url: examples/jentic-execute-example.json
- type: Examples
url: examples/jentic-register-example.json
- aid: jentic:jentic-search-api
name: Jentic Search API
description: Semantic search over the Jentic API and workflow catalog using natural language queries.
humanURL: https://docs.jentic.com/reference/public-apis/
baseURL: https://api.jentic.com/api/v1
tags:
- Search
properties:
- type: OpenAPI
url: openapi/jentic-search-api-openapi.yml
- type: Website
url: https://jentic.com/
- type: Documentation
url: https://docs.jentic.com/
- type: GettingStarted
url: https://docs.jentic.com/getting-started/quickstart/
- type: APIReference
url: https://docs.jentic.com/reference/public-apis/
- type: Signup
url: https://app.jentic.com/sign-up
- type: Console
url: https://app.jentic.com/
- type: BaseURL
url: https://api.jentic.com/api/v1
- type: SpectralRules
url: rules/jentic-rules.yml
- type: JSONSchema
url: json-schema/jentic-operation-schema.json
- type: JSONSchema
url: json-schema/jentic-agent-schema.json
- type: JSONSchema
url: json-schema/jentic-workflow-schema.json
- type: JSONStructure
url: json-structure/jentic-operation-structure.json
- type: Examples
url: examples/jentic-search-example.json
- type: Examples
url: examples/jentic-load-example.json
- type: Examples
url: examples/jentic-execute-example.json
- type: Examples
url: examples/jentic-register-example.json
- aid: jentic:jentic-access-requests-api
name: Jentic Access Requests API
description: 'Actor-agnostic, multi-item access-request surface (Core / Access bounded context). When an agent''s brokered
call is rejected by a permission rule — or the agent needs access it doesn''t yet have — the agent files an `AccessRequest`
containing one or more line items (`AccessRequestItem`). The request enters `pending` state and surfaces an `approve_url`
that the agent presents to a human reviewer.
Reviewers `:decide` individual items (approve or deny each); filers can `:amend` pending items (adjust rules or target)
or `:withdraw` the entire request. Each item transitions independently; the envelope status reflects the aggregate (`pending`
while any item is pending, `partially_approved` when some items are decided but others remain, terminal once all items
resolve).
Pending requests carry a TTL (default 7 days, configurable). Envelope lifecycle: `pending → partially_approved → approved
| denied | withdrawn | expired`.
**Identity scoping.** List and get operations are identity-scoped: filers see their own requests, reviewers see their
inbox, `org:admin` sees all. Non-owners receive `404` (not `403`) to avoid leaking existence.
**Security.** Mutating operations (`POST`, `:decide`, `:amend`, `:withdraw`) require `agents:write`; read operations require
`agents:read`.'
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Access Requests
properties:
- type: OpenAPI
url: openapi/jentic-access-requests-api-openapi.yml
- aid: jentic:jentic-actors-api
name: Jentic Actors API
description: Unified actor directory — a lightweight read-only view across all actor types (users, agents, service accounts).
Returns ID-to-name mappings for UI cache hydration so dashboards can display friendly names wherever an `actor_id` appears.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Actors
properties:
- type: OpenAPI
url: openapi/jentic-actors-api-openapi.yml
- aid: jentic:jentic-agent-registration-api
name: Jentic Agent Registration API
description: Dynamic agent registration (RFC 7591-style) — register a new agent client, poll registration status, update,
or delete a registration.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Agent Registration
properties:
- type: OpenAPI
url: openapi/jentic-agent-registration-api-openapi.yml
- aid: jentic:jentic-agents-api
name: Jentic Agents API
description: Agent actors — autonomous principals that call the Broker. Covers the agent lifecycle (list, read, approve
/ deny, enable / disable, archive) and the toolkits bound to each agent.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Agents
properties:
- type: OpenAPI
url: openapi/jentic-agents-api-openapi.yml
- aid: jentic:jentic-api-operations-api
name: Jentic API Operations API
description: 'Sub-resource of **APIs** (Registry bounded context). Enumerate the operations exposed by a registered API
— either the live (`current_revision_id`) view or a specific revision. Complements `Search` (ranked discovery) and `Inspect`
(single-operation detail) by giving callers the deterministic list of every registered operation under one API without
ranking, paging through `Search`, or parsing the spec.
Each row is slim — the structural detail (parameters, response schema, security scheme) lives behind `_links.inspect`
for callers that need it.'
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- API Operations
properties:
- type: OpenAPI
url: openapi/jentic-api-operations-api-openapi.yml
- aid: jentic:jentic-api-spec-api
name: Jentic API Spec API
description: 'Sub-resource of **APIs** (Registry bounded context). Download the underlying OpenAPI document for a registered
API — either the live (`current_revision_id`) view or a specific revision. Supports `application/json`, `application/openapi+yaml`,
and `application/yaml` via `Accept`-header content negotiation, with `Content-Disposition: attachment; filename="..."`
so browser downloads work cleanly.
By default the response includes any deployment-local overlays merged on top of the imported base. Pass `?overlays=false`
to download the raw imported spec instead.'
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- API Spec
properties:
- type: OpenAPI
url: openapi/jentic-api-spec-api-openapi.yml
- aid: jentic:jentic-apis-api
name: Jentic APIs API
description: 'The `Api` aggregate root — what APIs are registered for this deployment, addressable by the loose `(vendor,
name, version)` tuple. Every successful import lands a new immutable `ApiRevision`; the `Api` carries the live `current_revision_id`
plus presentation metadata (`display_name`, `description`, `icon_url`).
Mutation surfaces:
- `POST /apis` — async import (url / inline). Returns 202 + `Job (kind=import)`.
- `PATCH /apis/{vendor}/{name}/{version}` — metadata-only updates (display name, description, icon). Cannot mutate revisions
or spec content.
- `POST /apis/{vendor}/{name}/{version}/revisions/{revision_id}:promote` — make a draft revision live.
- `POST /apis/{vendor}/{name}/{version}/revisions/{revision_id}:archive` — retire a draft revision.
- `DELETE /apis/{vendor}/{name}/{version}` — deregister the API and all its revisions.
Newly imported revisions are **never auto-promoted** — callers must explicitly `:promote` to flip the live pointer. Hot-path
Broker calls can pin to a specific revision via the `Jentic-Revision` request header for testing without flipping the
pointer.
Sub-resources of an `Api` live under sibling tags: `API Operations` enumerates the operations exposed by a registered
revision; `API Spec` returns the underlying OpenAPI document for tooling.'
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- APIs
properties:
- type: OpenAPI
url: openapi/jentic-apis-api-openapi.yml
- aid: jentic:jentic-audit-api
name: Jentic Audit API
description: Append-only administrative audit trail — who did what, when. Distinct from `Executions` (brokered upstream
calls); this records control-plane mutations.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Audit
properties:
- type: OpenAPI
url: openapi/jentic-audit-api-openapi.yml
- aid: jentic:jentic-catalog-api
name: Jentic Catalog API
description: 'The public API **catalog** — a registry-side browse/preview/import surface over the Jentic public-APIs GitHub
manifest. Powers the Discover experience: list importable APIs (`GET /catalog`, keyset cursor-paginated), preview an entry''s
operations (`GET /catalog/{api_id}/operations`), refresh the cached manifest (`POST /catalog:refresh`, `org:admin`), and
import an entry into the local registry (`POST /catalog/{api_id}:import`).
The catalog is a cache of an upstream document, distinct from the local registry (`GET /apis`): `/apis` is what this deployment
has imported, `/catalog` is what it *could* import. Each entry''s `registered` flag reflects whether its `spec_url` already
backs a local API revision. Import resolves to a plain fetchable spec URL and reuses the standard async import job — no
catalog identity ever crosses into the importer.'
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Catalog
properties:
- type: OpenAPI
url: openapi/jentic-catalog-api-openapi.yml
- aid: jentic:jentic-configuration-api
name: Jentic Configuration API
description: Runtime, DB-backed platform configuration. Lets an operator set, read, and list configuration that previously
required hand-editing backend YAML and restarting the server — starting with credential provider configs (e.g. Pipedream).
A successful write rebuilds the in-process provider registry so the change takes effect without a restart. Secret fields
(e.g. `client_secret`) are encrypted at rest and redacted on read. **Topology note:** in the combined (single-process)
deployment a write takes effect immediately for all surfaces. In a multi-process deployment the control/broker processes
pick up the new config at their next boot, not the instant the admin write lands; cross-process propagation is a tracked
follow-up.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Configuration
properties:
- type: OpenAPI
url: openapi/jentic-configuration-api-openapi.yml
- aid: jentic:jentic-credentials-api
name: Jentic Credentials API
description: Part of the **Core / Access** bounded context — the authoritative record of "who can call what, with which
secret". A `Credential` is the auth secret + header configuration the Broker injects on the way out to an upstream API.
Credentials reference registry content via the loose `(vendor, name, version)` identity tuple (see `APIReference`) — there
is no foreign key between Core and Registry, by design. Cleartext secret material is returned **exactly once** at creation
(and on rotation via `PATCH`). Read paths return a redacted projection — last-N characters, hints, never the secret itself.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Credentials
properties:
- type: OpenAPI
url: openapi/jentic-credentials-api-openapi.yml
- aid: jentic:jentic-discovery-api
name: Jentic Discovery API
description: Unauthenticated metadata discovery — the JWKS document and OAuth authorization server metadata under `/.well-known/*`.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Discovery
properties:
- type: OpenAPI
url: openapi/jentic-discovery-api-openapi.yml
- aid: jentic:jentic-events-api
name: Jentic Events API
description: Curated, severity-tagged event stream surfaced to dashboards and operators. Each event is either **informational**
(no action required, e.g. `import.completed`) or **actionable** (requires operator follow-up, e.g. `credential.expiring_soon`,
`execution.repeated_failure`). Events reference the underlying `ExecutionRecord` or `Job` via `_links` and share `trace_id`
for correlation. Subscribe live via `GET /events/stream` (Server-Sent Events) or poll `GET /events` with a `since=` filter.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Events
properties:
- type: OpenAPI
url: openapi/jentic-events-api-openapi.yml
- aid: jentic:jentic-executions-api
name: Jentic Executions API
description: Append-only audit log of every Broker call (sync or async) — timing, status, trace IDs, and the upstream API
reference (`vendor:name:version`). Bodies are not stored. Records are written by the Broker; this surface is read-only.
To walk the history of a single logical request that fanned out into multiple upstream calls, list with `?trace_id={trace_id}`.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Executions
properties:
- type: OpenAPI
url: openapi/jentic-executions-api-openapi.yml
- aid: jentic:jentic-identity-api
name: Jentic Identity API
description: Identity introspection for the calling principal (human, agent, or service account) — `GET /me` returns the
resolved subject, scopes, and permissions behind the presented token.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Identity
properties:
- type: OpenAPI
url: openapi/jentic-identity-api-openapi.yml
- aid: jentic:jentic-inspect-api
name: Jentic Inspect API
description: Resolve a single registered operation to its full structural detail — parameters, response schema, security
scheme, server URL. Heavier than `Lookup` (which is the Broker hot path) and separate from `Search` (which only returns
ranked summaries). Supports JSON, Markdown, and OpenAPI YAML response formats via `Accept`-header content negotiation.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Inspect
properties:
- type: OpenAPI
url: openapi/jentic-inspect-api-openapi.yml
- aid: jentic:jentic-jobs-api
name: Jentic Jobs API
description: In-flight async-job tracker for both Registry import jobs and Broker async-execution jobs — one uniform `Job`
shape keyed by `kind`. Lifecycle only — type-specific result payloads (the imported revisions for `import`, upstream response
body for `execution`) live under `/jobs/{job_id}/result`. Result availability follows the organisation-level retention
policy (one-shot or TTL); once that expires, both `/jobs/{job_id}` and `/jobs/{job_id}/result` `404`.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Jobs
properties:
- type: OpenAPI
url: openapi/jentic-jobs-api-openapi.yml
- aid: jentic:jentic-mcp-api
name: Jentic MCP API
description: MCP (Model Context Protocol) transport reporting. The `jentic mcp` stdio server terminates all MCP protocol
traffic locally; the control plane only sees plain HTTP. This tag covers the small reporting surface behind it — today,
the config-registration report `jentic setup`/`jentic skill init` send after writing an MCP server entry for a detected
agent runtime, which feeds the config-written → first-session → first-execute adoption funnel.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- MCP
properties:
- type: OpenAPI
url: openapi/jentic-mcp-api-openapi.yml
- aid: jentic:jentic-monitoring-api
name: Jentic Monitoring API
description: Dashboard-oriented aggregation endpoints — execution volume, success/failure ratios, and top operations over
a bounded time window. Results are cached in-process (TTL ~120 s) so concurrent dashboard viewers never stampede the database.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Monitoring
properties:
- type: OpenAPI
url: openapi/jentic-monitoring-api-openapi.yml
- aid: jentic:jentic-notes-api
name: Jentic Notes API
description: 'Free-form annotations attached to registry resources — an `Api`, an `ApiRevision`, a registered `operation_id`,
an `ExecutionRecord`, or a `Credential`. Used to capture human or agent observations about quirks: *"this endpoint silently
truncates email addresses to 64 chars"*, *"this credential rejects calls outside US business hours"*, *"this operation
needs `Accept: application/json` even though the spec doesn''t say so"*.
Notes are an information surface for agents — they accumulate over time as the platform learns from real executions and
surface back through `_links.notes` on the resources they annotate (woven in by individual surfaces as they need it).
**Edit model.** Notes are editable via `PATCH`; each successful edit bumps a monotonic `revision` counter on the `Note`.
Use the `If-Match` header on `PATCH` and `DELETE` to guard against concurrent modification.
**Confidence promotion.** Notes carry both a `confidence` value (`observed | suspected | verified`) and a `confidence_source`
(`client | platform_promoted`). Today only `client` is written by callers; the `platform_promoted` value is reserved for
a future surface that auto-promotes `observed` notes to `verified` once enough matching `execution_feedback` accumulates.
**Authentication.** Any authenticated principal (agent or human) can post any note type. Per-type policy (e.g. only humans
may file `correction`) is enforced platform-side and not modelled in the spec.'
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Notes
properties:
- type: OpenAPI
url: openapi/jentic-notes-api-openapi.yml
- aid: jentic:jentic-oauth-api
name: Jentic OAuth API
description: OAuth 2.0 / OIDC endpoints exposed by the platform authorization server — the authorize, token, introspection,
revocation, and assertion-mint endpoints plus the redirect callback.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- OAuth
properties:
- type: OpenAPI
url: openapi/jentic-oauth-api-openapi.yml
- aid: jentic:jentic-oauth-clients-api
name: Jentic OAuth Clients API
description: Admin-managed registry of third-party OAuth clients (confidential, secret-bearing). Registered clients integrate
with Jentic One via the standard Authorization Code + PKCE flow. Admins can create, list, update, rotate secrets, and
deactivate clients. Deactivating a client immediately invalidates all tokens issued through it.
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- OAuth Clients
properties:
- type: OpenAPI
url: openapi/jentic-oauth-clients-api-openapi.yml
- aid: jentic:jentic-overlays-api
name: Jentic Overlays API
description: 'OpenAPI Overlay 1.0 (`https://spec.openapis.org/overlay/v1.0.0`) documents that correct an imported spec without
rewriting it. Overlays capture wrong required fields, missing security schemes, undocumented headers, silent payload truncation,
and other places where the upstream spec disagrees with the upstream API''s actual behaviour — without forking the spec
or waiting for an upstream fix.
Overlays attach to the `Api` aggregate (not to a specific `ApiRevision`) and continue to apply across revision bumps until
materially superseded. The optional `target_revision_id` records the revision the overlay was authored against, for auditing.
**Lifecycle.** New overlays start in `pending`. The Broker calls `POST /apis/{vendor}/{name}/{version}/overlays/{overlay_id}:confirm`
after the first successful upstream execution that exercised an overlay-patched operation, transitioning the overlay to
`confirmed`. This is the platform''s only documented data-plane → control-plane writeback. Once `confirmed`, an overlay
stays `confirmed` — repeated `:confirm` calls return `200` with the existing record unchanged.
**Edit model.** Overlay documents are editable via `PATCH` only while `pending`. Once `confirmed` (or `deprecated`), the
document is immutable — to change a confirmed overlay, submit a new one and let the old one age out via `DELETE` (which
soft-deprecates).
**Deletion.** `DELETE` is a **soft-deprecation** — the row is preserved with `status: deprecated` for audit. Deprecated
overlays no longer apply to spec-download merges (`?overlays=false` becomes the default for that overlay) but stay queryable
via `GET`. There is no hard-delete in this surface.
**Authentication.** Any authenticated principal (agent or human) can submit, edit, confirm, or deprecate overlays. Platform-side
rate limits gate abuse.'
humanURL: https://docs.jentic.com/guides/mcp/remote-mcp/
baseURL: https://api.jentic.com/mcp
tags:
- Overlays
properties:
- type: OpenAPI
url: openapi/jentic-overlays-api-openapi.yml
# --- truncated at 32 KB (43 KB total) ---
# Full source: https://artifacts.apis.io/jentic/apis.yml
Jentic API — frequently asked questions
Does Jentic have an API?
Yes. APIs.io lists 39 APIs from Jentic, including Jentic Remote MCP Server, Arazzo Engine and Jentic Standard Agent, and 36 more.
Is there an OpenAPI specification for the Jentic API?
Yes. APIs.io indexes 34 OpenAPI specifications for Jentic, each linked for download from this page.
How do I authenticate with the Jentic API?
Jentic uses bearer-token authentication (apiKey · 1 scheme). The details are profiled on APIs.io as Jentic Authentication.
How much does the Jentic API cost?
Jentic publishes its prices; its Jentic One Enterprise plan is Usage-based as of 2026-10-28. APIs.io lists 3 pricing plans for Jentic, with each plan's price, limits and inclusions under Pricing on this page.
Does Jentic have an MCP server?
Yes. APIs.io lists 1 MCP server for Jentic: Remote MCP server at api.jentic.com over HTTP; 4 tools listed.
Is the Jentic API ready for AI agents?
Jentic has a Kin Score of 57.9 out of 100 (strong) and an Agent Readiness score of 45.9 (agent-ready), scored by API Evangelist on 2026-11-06.
Calls without X-API-Key need no key. Calls with it need a paid plan — get a key.
Get an API key
Free tier, no form to fill in. Signing in shares your email address with us — we
store it to create your key and to recognise you if you sign in with another
provider. See our Privacy Policy and
Terms.