Red build on main
A developer asks why main is red. Memtro finds the latest pipeline on main, the failed workflow and job, and returns the failing tests with their messages and a link to the job.
Why the build failed, which tests broke and how often main goes red.
Connect CircleCI with a personal API token and Memtro lists a project's recent pipelines with their workflows, the jobs in a workflow, one job with its failed tests, and per-workflow insights such as success rate and credits. With a default project set, the hourly sync writes each failed workflow into memory. When you ask, and once an administrator has granted each action by name, it can trigger pipelines, rerun or cancel workflows and jobs, approve a waiting job, manage scheduled pipelines and roll a component back. Triggering, rerunning, approving and rolling back use credits and can deploy to production.
Ask in plain English from chat, your editor, a scheduled job or your own agent. The model picks the right CircleCI tool, reads what it needs, and answers with the source cited.
These tools change something in CircleCI. 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, choose the kinds you want, and Memtro registers its address with CircleCI for you. Each event can start a job, with the event as its input. CircleCI signs every delivery, and one whose signature does not check out is refused before it is read.
Events and webhooksOnce CircleCI 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.
Did the last build on main pass?
Why did job 4817 fail, and which tests broke?
What is the success rate of the deploy workflow over the last 30 days?
How many credits did the nightly workflow use last week?
Rerun the failed jobs in the last workflow on main
A developer asks why main is red. Memtro finds the latest pipeline on main, the failed workflow and job, and returns the failing tests with their messages and a link to the job.
Owen Price sees that main failed on a test that times out now and then, and asks Memtro to rerun only the failed jobs. Memtro reruns the workflow from its failed jobs in CircleCI as a new run, which uses credits, and reports back when it passes.
The teams that lean on CircleCI most, a question each of them asks, and what they connect beside it.
A record in CircleCI 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.
In app.circleci.com open User settings → Personal API Tokens → Create New Token. It acts as you, so it sees and runs every project you can. Set an expiry: CircleCI emails before it runs out
In Memtro open Connections → CircleCI, paste the Personal API token with a Default project slug such as gh/acme/web (or the project's app.circleci.com URL). The hourly sync of failed workflows needs one
Self-hosted CircleCI server? Add its root URL as the Base URL. Connect, and Memtro checks the token and the project
Ask whether the last build on main passed
Chat in the dashboard, the OpenAI-compatible API, scheduled jobs, and any MCP client: Claude Code, OpenCode, Cursor, Claude Desktop. The CircleCI tools are the same everywhere.
Only when you ask, and only what it has been granted: trigger a pipeline on a branch or tag (uses credits); rerun a workflow, or only its failed jobs (uses credits); cancel a running workflow or job; approve a job waiting on approval, such as a deploy; create, change or delete a scheduled pipeline; roll a deployed component back to an earlier version. 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. An hourly sync anchors recently changed CircleCI records into memory as entities.
Personal API token. In app.circleci.com open User settings → Personal API Tokens → Create New Token. It acts as you, so it sees and runs every project you can. Set an expiry: CircleCI emails before it runs out.
Yes. Turn events on in the connection's Events tab on the Connections page, choose the kinds you want, and Memtro registers its address with CircleCI for you. 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