Who deployed what to Heroku and when, and which dynos are running or crashed.
Connect Heroku with your API key or an authorization token and Memtro reads apps with region, stack, team and URL, releases with who made them, builds, running dynos and process formation, collaborators, custom domains and teams. Config vars and logs are never read. It reads on demand and there is no hourly sync. When you ask, and once an administrator has granted the action by name, it can restart and scale dynos, deploy, roll back, cancel builds, run one-off commands, change or create apps, manage collaborators and domains, and permanently delete an app. Scaling and one-off dynos spend money.
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right Heroku tool, reads what it needs, and answers with the source cited.
These tools change something in Heroku. The model is told to use them only when you ask for exactly that, and an administrator grants each one by name to the teams that may use it; a personal connection acts only after its owner allows it.
Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in Heroku, choosing the same kinds; the tab shows Heroku's own steps. Each event can start a job, with the event as its input. Heroku signs every delivery, and one whose signature does not check out is refused before it is read.
Events and webhooksOnce Heroku is connected, these are the kinds of things people ask. Plain questions, no syntax. If the answer lives partly in another system you have connected, Memtro looks there too.
What was deployed to the billing app this week, and by whom?
Are any dynos crashed on the API app right now?
Which builds of the web app failed this week, and who started them?
Who has access to the production app?
Roll the API app back to the release before this morning's deploy
Errors started at noon and the engineering lead asks what changed on the API app. Memtro reads Heroku's releases newest first, with each version's description, who made it and when, and points to the release just before the errors began.
Dan Fletcher asks Memtro to roll the API app back after the noon release broke checkout. Memtro names the current and earlier versions, and once he confirms, Heroku creates a new release running the earlier code and config and restarts the app on it.
The teams that lean on Heroku most, a question each of them asks, and what they connect beside it.
A record in Heroku is rarely the whole story. Connect the systems that hold the rest and Memtro checks them together, then remembers what it learned so the next question starts further along.
Using something else for the same job, or both? Each connector has its own tools; connect as many as you like.
UptimeRobot
Honeycomb
Cronitor
FireHydrant
StatusCake
GlitchTip
GLPI
Langfuse
LibreNMS
NetBox
Snipe-IT
Woodpecker CI
Zabbix
LangSmith
Helicone
GrowthBook
Drata
Intruder
GitGuardian
Secureframe
ilert
JumpCloud
ConfigCat
Split
Aikido Security
Astronomer
Comet Opik
CrowdStrike Falcon
Detectify
Domotz
LambdaTest
LangWatch
LlamaCloud
Logz.io
Lunary
NinjaOne
Northflank
Portkey
PromptLayer
PRTG Network Monitor
Raygun
Retell AI
Runpod
Shodan
Spike.sh
Squadcast
Tenable
UpGuard
Uptime.com
Vanta
Zenduty
Cortex
WorkOS
Kandji
Automox
Huntress
SimpleMDM
Healthchecks.io
Rapid7 InsightVM
ZenhubIn the Heroku Dashboard → Account settings → API Key, reveal the key and copy it, though a dedicated Heroku user added only to the apps Memtro should operate is safer than your own key
Or run heroku authorizations:create in the Heroku CLI for a token that is not your account key, and copy the token it prints
In Memtro open Connections → Heroku, paste it as the API key or authorization token and connect
Ask what was deployed this week, noting Heroku allows about 75 API calls a minute per account, shared with your own tools
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The Heroku tools are the same everywhere.
Only when you ask, and only what it has been granted: restart all of an app's dynos, one dyno, or one process type; scale a process type: change how many dynos it runs or their size (spends money); roll an app back to an earlier release; deploy an app from a source tarball URL (build and release); cancel a running build; run a one-off command in a new dyno (spends money, runs code against the app's data); turn maintenance mode on or off, rename an app, or change its build stack; create an app; delete an app permanently; add a collaborator to an app (invitation email off unless asked), or remove one; attach a custom domain to an app. Every other tool reads, searches or lists. The Connections page labels the tools that act, an administrator grants each one by name to the teams that may use it, and a personal connection acts only after its owner allows it.
A personal connection serves only your own requests. A shared connection serves the organisation: everyone, or only the teams an admin chooses. Admins can always add shared connections; other members need the permission.
Credentials are encrypted at rest. Tool results are cached briefly so agent loops do not repeat identical calls, and the facts the model extracts are saved to your organisation's memory with provenance (system, record and link) so they can be checked and superseded.
API key or authorization token. In the Heroku Dashboard → Account settings → API Key, reveal the key and copy it, though a dedicated Heroku user added only to the apps Memtro should operate is safer than your own key.
Yes. Turn events on in the connection's Events tab on the Connections page, then add the address it shows as a webhook in Heroku, choosing the same kinds; the tab shows Heroku's own steps. A job set to start "When an event arrives" then runs once for each matching event, with the event as its input. The run treats the event as outside data, never as instructions, and anything the run remembers is kept as unverified.
Seven-day free trial, then one monthly price per organisation. Connectors, jobs and memory are never metered.
Sign in, add a provider key, create an API key and point your tools at Memtro. It takes about five minutes.
7-day free trial · card on file, nothing charged until it ends · cancel any time
