Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the RevDesk v1 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%2Frevdesk.com%2Frevdesk" | 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%2Frevdesk.com%2Frevdesk" | 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 RevDesk v1 API.
Place a single-leg outbound AI call to a phoneNumber with POST /v1/calls, optionally choosing the assistantId and script
Bridge your from_number to a destination with POST /v1/calls/dial, with record and amd settings and an optional connect_number for the human leg
Send an SMS from an owned number with POST /v1/sms/send using to, from, and message
Search available numbers by country, area_code, and number_type, then purchase one with POST /v1/phone-numbers
GET STARTED
Enroll numbers in reputation monitoring and read Hiya-sourced spam risk scores for a number from /v1/caller-trust/reputation/numbers/{phone}
Hand RevDesk an outcome with POST /v1/tasks, giving an objective, deadline, success_criteria, and guardrails such as quiet_hours and max_touches
Read a finished call's transcript from /v1/calls/{id}/transcript and subscribe a webhook_url to events such as call_completed or sms_delivered
Patterns agents use RevDesk v1 API for, with concrete tasks.
★ Agent-placed outreach calls through Jentic
A sales agent connected through Jentic searches by intent for placing an AI phone call, gets POST /v1/calls with its schema, and dispatches a call to the lead's phoneNumber with a short script. When the call completes, the agent reads GET /v1/calls/{id}/transcript and writes the outcome into the CRM, so the next step is based on what was actually said.
POST /v1/calls with phoneNumber '+14155550123', name 'Jordan', and script 'Follow up on the pricing proposal', then GET /v1/calls/{id}/transcript once the call has ended
Number reputation and spam-label remediation
Outbound numbers that carriers label as spam stop getting answered. An agent enrolls the team's numbers with POST /v1/caller-trust/reputation/numbers, reads each number's spam risk and component scores, and submits flagged numbers for remediation with POST /v1/caller-trust/reputation/remediations. Cached reads are free, while fresh=true forces a live billed carrier query.
POST /v1/caller-trust/reputation/numbers with phone_numbers ['+14155550123'], then GET /v1/caller-trust/reputation/numbers/+14155550123 with fresh false and submit it to /v1/caller-trust/reputation/remediations if the risk is high
Outcome tasks with guardrails
Instead of scripting every touch, POST /v1/tasks hands RevDesk an objective, such as booking a meeting with a contact_id, plus a deadline, success_criteria outcome_kinds, and guardrails covering channels, quiet_hours, and max_touches. RevDesk works across calls and SMS until the criteria are met, the deadline passes, or a guardrail stops it, and GET /v1/tasks/{taskId} reports progress.
POST /v1/tasks with type 'outreach', objective 'Book a demo with this contact', contact_id 'c_123', and guardrails {channels: ['sms','call'], quiet_hours: {from: 20, to: 8, time_zone: 'America/Los_Angeles'}, max_touches: 3}
Number provisioning and SMS auto-reply
An operations agent searches available numbers with GET /v1/phone-numbers/search by country, area_code, and number_type, buys one with POST /v1/phone-numbers, and then sets its CNAM display name, recording, and assigned agent_id with PATCH /v1/phone-numbers/{id}. A per-number SMS auto-reply override can be set with reply_delay_seconds and a default_prompt.
GET /v1/phone-numbers/search with country 'US', area_code '415', number_type 'local', limit 5, then POST /v1/phone-numbers with the chosen phone_number and name 'SF sales line'
119 endpoints — the revdesk v1 api places outbound ai voice calls and human bridge calls, sends sms, and manages the phone numbers, caller ids, and caller reputation behind them for a sales or service team.
METHOD
PATH
DESCRIPTION
/v1/calls
Place an outbound AI call
/v1/calls/dial
Bridge two phone numbers
/v1/sms/send
Send an SMS message
/v1/calls/{id}/transcript
Retrieve a call transcript
/v1/phone-numbers/search
Search available phone numbers
/v1/caller-trust/reputation/numbers/{phone}
Get reputation scores for a number
/v1/tasks
Create an outcome task
/v1/webhook_subscriptions
Create a webhook subscription
/v1/calls
Place an outbound AI call
/v1/calls/dial
Bridge two phone numbers
/v1/sms/send
Send an SMS message
/v1/calls/{id}/transcript
Retrieve a call transcript
/v1/phone-numbers/search
Search available phone numbers
/v1/caller-trust/reputation/numbers/{phone}
Get reputation scores for a number
/v1/tasks
Create an outcome task
/v1/webhook_subscriptions
Create a webhook subscription
What agents get from Jentic-routed access to this vendor.
Setup
Wiring RevDesk by hand means managing an organization-scoped bearer key with the right x-required-scopes for each area, sending Idempotency-Key headers on writes, and keeping legacy deprecated paths apart from their replacements. Through Jentic you install once, import the RevDesk v1 API from the API Directory, store the key once, and your agent calls it.
Permission scoping
RevDesk puts phone number, agent, and call ids in the URL path (/v1/phone-numbers/{id}, /v1/agents/{id}, /v1/calls/{id}), so a rule can pin your agent to one phone number's settings or one AI agent's configuration. You choose the operations it may call, so actions like releasing a number with DELETE /v1/phone-numbers/{id} or placing calls are not included unless you add them.
Credential isolation
Your RevDesk 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 'place an AI phone call' or 'check if my number is flagged as spam', and Jentic returns the matching RevDesk 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 RevDesk v1 API through Jentic.
What authentication does the RevDesk v1 API use?
Per its OpenAPI spec, the API takes an organization-scoped RevDesk API key as a bearer token in the Authorization header, and each operation declares the permission it needs in x-required-scopes, such as calls:write, sms:write, or caller_trust:read. The sandbox endpoints at /v1/sandbox need no auth and never read tenant data. Through Jentic the key is stored encrypted by your own Jentic One instance and injected at call time.
What is the difference between POST /v1/calls and POST /v1/calls/dial?
POST /v1/calls dispatches a single-leg outbound AI call: an AI agent, chosen by assistantId or your most recently created agent, joins and speaks, using one channel. POST /v1/calls/dial bridges your from_number to a destination for a human call, using two concurrent channels, and requires to_number, record, and amd.
Which SMS endpoint should I use with the RevDesk v1 API?
Use POST /v1/sms/send with to, from, and message, where from must be an active number your organization owns. The older POST /v1/messages is marked deprecated in the spec, as are the legacy /v1/jobs paths, which /v1/tasks replaces.
What are the rate limits for the RevDesk v1 API?
The OpenAPI spec does not give a general rate limit. It does state that a fresh reputation read on GET /v1/caller-trust/reputation/numbers/{phone} is a billed live carrier query limited to one per day per number, while cached reads are free. Check https://docs.revdesk.com for current limits.
Is there a RevDesk MCP server?
You don't need an MCP server to give your agent the RevDesk v1 API. Jentic connects it directly from the API Directory: import it, store your API key once, and your agent calls it. Operations are found by intent when needed, so no extra tool definitions sit in your agent's context.
How do I place an AI phone call through Jentic?
Install Jentic One, add the RevDesk v1 API from the directory, and store an API key with the calls:write permission. Your agent searches for 'place an outbound AI phone call', gets POST /v1/calls with its schema, and calls it with the phoneNumber and an optional script. To run it on your own infrastructure, install Jentic One from its GitHub repo.
Can I limit what my agent is allowed to do with the RevDesk v1 API?
Yes. Write a rule that allows only GET /v1/calls, GET /v1/calls/{id}, and GET /v1/calls/{id}/transcript, so the agent can review calls and read transcripts but cannot dial, send SMS, or buy numbers, and every call it makes is logged.
For Agents
Place outbound AI or bridged calls, send SMS, buy and configure phone numbers, check number spam reputation, read call transcripts, and hand RevDesk tasks to pursue across channels.
Use for: Call this lead with our AI agent and introduce the new pricing, I need to send an SMS reminder from our sales number, Check whether our outbound number is being flagged as spam, Find a local phone number in area code 415 and buy it
Not supported: Does not handle email campaigns, CRM deal pipelines, payments, or video meetings. Use for AI and bridged phone calls, SMS, and the phone numbers and caller reputation behind them only.
The RevDesk v1 API places outbound AI voice calls and human bridge calls, sends SMS, and manages the phone numbers, caller IDs, and caller reputation behind them for a sales or service team. It also covers AI agent configuration, contacts, call transcripts, outcome-driven tasks that RevDesk pursues across channels, webhook subscriptions for call and SMS events, number health testing, CNAM and Free Caller Registry registration, and usage. An organization-scoped API key authorises each operation against the permission it declares, such as calls:write or sms:read.
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>