Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Recreate 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%2Fprosights.co%2Fprosights" | 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%2Fprosights.co%2Fprosights" | 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 Recreate API.
Upload a PDF or image-based file and choose the pages to extract with selected_pages, such as 1-11,12,14
Turn table extraction, chart extraction, and PPT processing on or off per request
Merge tables that span several pages into one with table_clustering_mode set to document
Poll a recreate job's status by recreate_id until it reports SUCCESS or ERROR
GET STARTED
Read the extracted tables and charts for a job as JSON, returned as they become available
Patterns agents use Recreate API for, with concrete tasks.
★ Agent-Driven Table Extraction
An AI agent connected through Jentic takes a PDF a user shares, submits it to the Recreate API with the pages that hold the financial tables, polls the job status, and then reads the tables back as JSON. The agent can then answer questions about the figures or load them into a spreadsheet, without anyone retyping numbers from the document.
Upload annual-report.pdf with selected_pages 12-18 and is_xls_process_tables_enabled true, poll the status by recreate_id, then read the tables with the to_json operation
Financial Statement Digitization
Analysts receive financial statements and pitch materials as PDFs or scans. The Recreate API pulls the balance sheet and income statement tables out as structured data, and setting table_clustering_mode to document keeps a statement that runs over several pages together as one table, so downstream models read one clean grid.
Submit a scanned statement with selected_pages 1-4 and table_clustering_mode document, then return the merged table from the JSON output
Chart Data Recovery
Slides and reports often carry charts with no underlying data attached. With chart processing turned on, the Recreate API returns detected charts alongside tables in the JSON output, so the series behind a bar or line chart can be reused in a new analysis.
Upload deck.pdf with selected_pages 5,9 and is_xls_process_charts_enabled true, then read the charts array from the JSON output
3 endpoints — the prosights recreate api extracts tables and charts from pdf or image-based documents and returns them as structured data.
METHOD
PATH
DESCRIPTION
/enterprise-api/recreate/
Submit a document for table and chart extraction
/enterprise-api/recreate/status/
Fetch a recreate job's status
/enterprise-api/recreate/{recreate_id}/to_json
Read extracted tables and charts as JSON
/enterprise-api/recreate/
Submit a document for table and chart extraction
/enterprise-api/recreate/status/
Fetch a recreate job's status
/enterprise-api/recreate/{recreate_id}/to_json
Read extracted tables and charts as JSON
What agents get from Jentic-routed access to this vendor.
Setup
Wiring the Recreate API by hand means sending the `X-Prosights-Api-Key` header, building a multipart upload with page ranges and string true or false flags, and polling a status operation before reading the JSON output. Through Jentic you install once, import the Recreate API from the API Directory, store the key once, and your agent calls it.
Permission scoping
The Recreate API puts the recreate_id in the URL path when reading results (/enterprise-api/recreate/{recreate_id}/to_json), so a rule can pin your agent to reading one job's output. You choose the operations it may call, so document submission is not included unless you add it.
Credential isolation
Your ProSights 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 'extract tables from a PDF' or 'check a table extraction job', and Jentic returns the matching Recreate 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 Recreate API through Jentic.
What authentication does the Recreate API use?
The Recreate operations authenticate with an API key sent in the `X-Prosights-Api-Key` header, under the APIKeyHeader scheme in the OpenAPI spec. Through Jentic the key is stored encrypted by your own Jentic One instance and injected at call time, so it never reaches the agent's prompt, logs, or context.
Which pages and options can I set when submitting a document?
The file and selected_pages fields are required, and selected_pages takes one-indexed page numbers or ranges such as 1-11,12,14,15. Optional fields turn chart processing, table processing, and PPT processing on or off, and table_clustering_mode chooses page or document clustering for tables.
What are the rate limits for the Recreate API?
The OpenAPI spec does not specify rate limits. Check https://docs.prosights.co for the current limits on your account.
Is there a ProSights Recreate MCP server?
You don't need an MCP server to give your agent the Recreate API. Jentic connects it directly from the API Directory: import it, store your credential once, and your agent calls the submit, status, and JSON output operations as it needs them.
Can I limit what my agent is allowed to do with the Recreate API?
Yes. For example, write a rule that allows only the status and to_json operations, so the agent can check on and read the results of documents you submitted yourself without uploading new files. Every call it makes is logged.
How do I extract tables from a PDF with the Recreate API through Jentic?
Import the Recreate API from the Jentic API Directory, store your API key once, and have your agent search Jentic for 'extract tables from a PDF'. Jentic returns the create recreate operation with its multipart input schema; the agent uploads the file with selected_pages, polls the status operation with the recreate_id, and reads the result from the to_json operation. To run it on your own infrastructure, install Jentic One from its GitHub repo.
For Agents
Upload a PDF or image document, pick the pages, and get its tables and charts back as JSON with the ProSights Recreate API.
Use for: Extract the tables from pages 3 to 7 of this PDF, Pull the charts out of this investor deck as data, Check whether my table extraction job has finished, Get the extracted tables as JSON for recreate job 8c2f
Not supported: Does not edit PDFs, fill forms, or extract invoice fields such as totals and vendor names. Use for extracting tables and charts from PDF and image documents only.
The ProSights Recreate API extracts tables and charts from PDF or image-based documents and returns them as structured data. You upload a file with the pages to process, written as one-indexed numbers or ranges such as 1-11,12,14, and choose whether table, chart, and PPT processing run. A table_clustering_mode of page or document decides whether a table that spans several pages comes back as one table. You then poll the job by recreate_id and read the extracted tables and charts as JSON, which is returned as it becomes available.
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>