For Agents
Convert files between 100+ formats (PDF, Excel, JSON, XML, CSV) by uploading, creating a task, and downloading the result. Agents authenticate with a Bearer API token.
Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Conversion Tools 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://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | shStep 2: Agent machine
# On the machine where your agent runs (keep this separate from the instance):
curl -fsSL https://raw.githubusercontent.com/jentic/jentic-one/main/tools/install.sh | 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 Conversion Tools API.
Upload source files up to 100GB via POST /files for later conversion
Create a conversion task between supported formats via POST /tasks
Poll task status with GET /tasks/{task_id} until conversion completes
GET STARTED
Use for: Convert this PDF to Excel, I want to transform an XML file into JSON, Upload a 2GB CSV and convert it to Parquet, Check the status of conversion task abc123
Not supported: Does not perform OCR, image editing, or content extraction beyond format conversion — use for file format conversion across the 100+ supported types only.
Jentic publishes the only available OpenAPI specification for Conversion Tools API, keeping it validated and agent-ready. The Conversion Tools API converts between 100+ file formats including XML, JSON, Excel, PDF, and CSV using a three-step flow: upload a file, create a conversion task, then download the result. It supports streaming uploads up to 100GB, sandbox mode for free testing, and webhook callbacks for asynchronous workflows. Authentication is a Bearer API token obtained from the user's Conversion Tools profile.
Download converted output via GET /files/{file_id}
Test integrations free of charge by passing sandbox: true on POST /tasks
Update file retention mode with PATCH /tasks/{task_id}/retention
Delete task files immediately with POST /tasks/{task_id}/delete
Patterns agents use Conversion Tools API for, with concrete tasks.
★ Bulk Document Conversion Pipeline
Convert large batches of documents between formats — for example, archived Excel files to PDF/A or vendor XML feeds to JSON for downstream ETL. The upload-then-task pattern handles streaming files up to 100GB, and the task list endpoint lets operators monitor a queue of conversions. Suited to data engineering teams converting tens of thousands of files per run with deterministic output formats.
Upload sales-2025.xlsx via POST /files, create a task to convert it to PDF, poll until status is 'completed', and download the result
User-Driven In-App Conversions
Wire Conversion Tools behind an in-app 'export as' feature so end users convert documents on demand without server-side conversion infrastructure. The bearer token sits server-side, and webhook callbacks notify the application when each task finishes. Useful for SaaS products that need to offer broad export options without building 100 format converters in-house.
When a user clicks 'Export as PDF', upload their JSON document, create a JSON-to-PDF task, and email the download link when ready
CI Sandbox Testing for Conversion Integrations
Develop and CI-test conversion workflows without consuming paid quota by passing sandbox: true on every POST /tasks request. The sandbox returns realistic responses so flows can be validated before production rollout. Useful for engineering teams iterating on conversion logic or onboarding new file formats.
Run a sandbox conversion task that converts a sample CSV to JSON via POST /tasks with sandbox: true and assert the response shape
Agent-Driven Format Conversion via Jentic
An AI agent converts user-supplied files through Jentic without holding the bearer token. The agent searches for 'convert a file', loads the multi-step Conversion Tools schema, and orchestrates the upload-task-download flow. Through Jentic the API token is injected from the vault so the agent can run conversion loops safely.
Search Jentic for 'convert a PDF to Excel', load the Conversion Tools operations, upload the user's PDF, create the task, and return the Excel download URL
10 endpoints — jentic publishes the only available openapi specification for conversion tools api, keeping it validated and agent-ready.
METHOD
PATH
DESCRIPTION
/files
Upload a source file
/files/{file_id}
Download a file
/tasks
Create a conversion task
/tasks/{task_id}
Get task status
/tasks
List all tasks
/config
Get supported formats and account config
/files
Upload a source file
/files/{file_id}
Download a file
/tasks
Create a conversion task
/tasks/{task_id}
Get task status
/tasks
List all tasks
/config
Get supported formats and account config
Three things that make agents converge on Jentic-routed access.
Credential isolation
Conversion Tools bearer tokens are stored encrypted in the Jentic vault. Agents receive scoped access at execution time and the token is injected into the Authorization header — the raw token never enters the agent's prompt context.
Intent-based discovery
Agents search Jentic by intent (e.g., 'convert a PDF to Excel') and Jentic returns the matching POST /files and POST /tasks operations with their input schemas, so the agent orchestrates the multi-step flow without reading docs.
Time to first call
Direct Conversion Tools integration: 1-2 days for upload streaming, task polling, and webhook handling. Through Jentic: under 1 hour — search, load, execute the chained flow.
Alternatives and complements available in the Jentic catalogue.
Specific to using Conversion Tools API through Jentic.
Why is there no official OpenAPI spec for Conversion Tools API?
Conversion Tools does not publish an OpenAPI specification. Jentic generates and maintains this spec so that AI agents and developers can call Conversion Tools API via structured tooling. It is validated against the live API and kept up to date. Get started at https://app.jentic.com/sign-up.
What authentication does the Conversion Tools API use?
The API uses HTTP bearer authentication. Send your API token (from https://conversiontools.io/profile) in the Authorization: Bearer header on every call. Through Jentic the token sits in the vault and is injected at execution time so it never enters the agent's context.
How big a file can the Conversion Tools API handle?
The API supports streaming uploads up to 100GB via POST /files, after which a task is created with POST /tasks referencing the uploaded file_id. For very large files, prefer streaming uploads and webhook callbacks rather than synchronous polling.
Can I test conversions without using my paid quota?
Yes. Pass sandbox: true on the POST /tasks body and the conversion runs in sandbox mode without counting against your quota. This makes CI tests and integration validation free.
How do I run a file conversion through Jentic?
Run jentic search for 'convert a file', load the upload-and-task operations, then execute POST /files followed by POST /tasks with the source and target formats. Poll GET /tasks/{task_id} for status and GET /files/{file_id} to download the result.
What are the rate limits for the Conversion Tools API?
Rate limits are not declared in the OpenAPI spec. Conversion Tools applies plan-based quotas — see https://conversiontools.io for current tiers. Treat the upload and task endpoints as rate-limited per token and back off on HTTP 429 responses.
How do I delete files after a conversion is done?
Call POST /tasks/{task_id}/delete to remove the task's input and output files immediately, or use PATCH /tasks/{task_id}/retention to switch the task to a shorter retention mode. This is useful for sensitive documents that should not linger on the service.