Install Jentic One Beta
Jentic One is a self-hosted execution layer for AI agents. It lets your agent call the Memchats 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%2Fmemchats.com%2Fmemchats" | 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%2Fmemchats.com%2Fmemchats" | 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 Memchats API.
Create a contact with a name and relationship_type such as client, cofounder, family, or dating, plus an optional context_note
Read a contact's conversation memory, a pre-call summary before a meeting, and a relationship health score
Ask Alma about a contact, letting her pick the area or choosing legal, relaciones, conocer, or negocio explicitly
Search message history by full-text query, optionally narrowed to one contact_id
GET STARTED
Import a person from dating-profile or chat screenshots and track the import until it finishes
Analyse dating-app screenshots into a profile and draft the next message in the user's own writing voice
Export a contact's full memory as downloadable JSON
Patterns agents use Memchats API for, with concrete tasks.
★ Pre-meeting briefings from an agent through Jentic
Before a call, a personal assistant agent pulls what Memchats remembers about the person on the calendar. Through Jentic it finds the contact, reads the pre-call summary and the health score, and asks Alma a focused question such as how to raise a delayed payment with a client. The mc_live_ key stays in your own Jentic One instance, and every Alma question draws from the plan's weekly LLM query quota.
Call GET /api/v1/contacts to find the contact named 'Laura Gómez', call GET /api/v1/contacts/{id}/pre-call and GET /api/v1/contacts/{id}/health-score for that id, then summarise both for the user
Advice about a specific person
Users ask Alma for help with one relationship at a time. The ask endpoint takes a contact_id and a question and routes to one of four areas (legal, relaciones, conocer, negocio); the respond endpoint lets Alma choose the area herself from who the contact is and what the exchange is about. Previous answers for a contact can be listed later, and feedback on a suggested reply records what the user actually did.
Call POST /api/v1/advisors/negocio/ask with contact_id set to the cofounder's id and question 'How should I propose changing our equity split?', then return Alma's answer
Dating conversation assistant
For dating contacts, the API analyses app screenshots into a profile, crops photos into a gallery, and drafts the next message with an intent and optional character limit. It can also translate stored texts and critique the user's own bio, rewriting it three times in their voice, so an assistant helps with both sides of a match.
Call GET /api/v1/dating to list dating conversations, pick the first one, then call POST /api/v1/contacts/{id}/suggest-message with intent 'propose_date' and return the draft
Personal memory search and export
Months of messages become searchable. The search endpoint runs a full-text query across conversations with a q parameter and optional contact_id, and the export endpoint returns everything Memchats keeps about a contact as downloadable JSON, so a user can find a past promise or take their data elsewhere.
Call GET /api/v1/search with q 'invoice' and the client's contact_id, list the matching messages, then call GET /api/v1/contacts/{id}/export for a JSON backup
77 endpoints — the memchats api keeps a conversation memory for each person in a user's life and lets alma, its counsellor, answer questions about that person.
METHOD
PATH
DESCRIPTION
/api/v1/contacts
Create a contact
/api/v1/contacts
List contacts
/api/v1/contacts/{id}/memory
Get contact memory
/api/v1/contacts/{id}/pre-call
Get pre-call summary
/api/v1/advisors/{slug}/ask
Ask Alma about a contact in one area
/api/v1/search
Full-text search in messages
/api/v1/contacts/{id}/suggest-message
Draft a dating message
/api/v1/imports/people
Import a person from screenshots
/api/v1/contacts
Create a contact
/api/v1/contacts
List contacts
/api/v1/contacts/{id}/memory
Get contact memory
/api/v1/contacts/{id}/pre-call
Get pre-call summary
/api/v1/advisors/{slug}/ask
Ask Alma about a contact in one area
/api/v1/search
Full-text search in messages
/api/v1/contacts/{id}/suggest-message
Draft a dating message
/api/v1/imports/people
Import a person from screenshots
What agents get from Jentic-routed access to this vendor.
Setup
Wiring Memchats by hand means generating an mc_live_ key, sending it as a bearer token, tracking asynchronous screenshot imports until they finish, and budgeting calls that draw on the weekly LLM quota. Through Jentic you install once, import Memchats from the API Directory, store the key once, and your agent calls it.
Permission scoping
Memchats puts the contact id in the URL path (/api/v1/contacts/{id}/...), so a rule can pin your agent to one contact and the operations you pick. You choose the operations it may call, so destructive ones like deleting a contact or a dating photo are not included unless you add them.
Credential isolation
Your Memchats 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 'summarise a contact before a call' or 'search my messages', and Jentic returns the matching Memchats 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 Memchats API through Jentic.
What authentication does the Memchats API use?
Per its OpenAPI spec, the Memchats API uses HTTP bearer auth with an API key that starts with mc_live_, generated in the app under API keys settings and sent in the Authorization header. Account endpoints such as signup, the token exchange at /api/v1/auth/token, and the sign-in provider routes do not require authentication. Through Jentic the key is stored encrypted by your own Jentic One instance and injected at call time.
Can I get a summary of a contact before a call with the Memchats API?
Yes. GET /api/v1/contacts/{id}/pre-call returns a pre-call summary for the contact, GET /api/v1/contacts/{id}/memory returns the full contact memory, and GET /api/v1/contacts/{id}/health-score returns a relationship health score, so an agent can brief the user in one pass.
What are the rate limits for the Memchats API?
The OpenAPI spec does not state numeric rate limits. It does note that Alma questions and conversation replies each consume one LLM query from the plan's weekly quota, and plans differ in how many analyses per week they allow. Check https://memchats.com/developers for current limits.
Is there a Memchats MCP server?
You don't need an MCP server to give your agent Memchats. Jentic connects it directly from the API Directory: import it, store your API key once, and your agent calls the contact, memory, and Alma operations it needs on demand without loading another server's tool definitions into its context.
How do I ask Alma about a contact through Jentic?
Search Jentic for 'ask Alma about a contact', load the POST /api/v1/advisors/{slug}/ask operation, and execute it with an area slug such as negocio, the contact_id, and your question. GET /api/v1/advisors lists the available areas first. 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 Memchats API?
Yes. For example, write a rule that allows only the pre-call, memory, and health-score reads under /api/v1/contacts/{id}/ for one client contact, so the agent can brief you on that person but cannot delete contacts, send messages, or ask Alma questions that use your weekly quota. Every call it makes is logged.
For Agents
Create Memchats contacts, read each contact's conversation memory, pre-call summary, and health score, search past messages, and ask Alma for advice about a person in the legal, relaciones, conocer, or negocio area.
Use for: I need a quick summary of my history with this client before our call, Search my messages for where we talked about the contract, Ask Alma how to handle a tense conversation with my cofounder, Check the relationship health score for my sister
Not supported: Does not send messages through WhatsApp, Telegram, or dating apps, manage sales pipelines, or book meetings. Use for personal relationship memory and Alma's advice about contacts only.
The Memchats API keeps a conversation memory for each person in a user's life and lets Alma, its counsellor, answer questions about that person. Contacts carry a relationship type such as friend, client, cofounder, or dating, and the API returns each contact's memory, a pre-call summary, a relationship health score, detected behaviour patterns, and a compatibility snapshot between two contacts. Alma answers across four areas (legal, relaciones, conocer, negocio), and a dating toolkit analyses profile screenshots, drafts messages, and critiques the user's own bio. Requests authenticate with an mc_live_ API key sent as a bearer token.
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>