- Industry
- IT Company
- Department
- Engineering
- Team
- Platform and Reliability
- Kind
- Investigator
- Approver
- Engineering Manager
- Code
- ENG-12
It diagnoses failed and slow builds and deployments across client and SaaS pipelines, telling a broken change from a flaky test, an expired credential, a runner shortage, a dependency outage or a client environment that changed underneath. It proposes the fix and records it on the pipeline run, never bypassing a protected gate such as a security scan or the client's release approval. **The problem it solves:** Why the build or deployment failed or slowed: the failing step, what changed before it, and which cause the evidence supports.
How it works
6 steps, and a person signs off before step 6.
- 01Looks it up
Finds the record the request is about in the source code repository and CI/CD: the pipeline run, its logs, recent commits and the runner and credential status..
- 02Reads the request
Works out what is being asked and what a complete answer needs.
- 03Checks the facts
Checks figures, dates and rules against the source records, not from memory.
- 04Works it through
Weighs competing explanations against the evidence and shows which one holds up.
- 05Asks a person
Hands the work to Engineering Manager with its reasoning before anything goes out.
- 06Acts
Once approved, it acts in the source code repository and CI/CD: it records the diagnosis on the run and retries or opens a fix. and records what was done.
What it works with
Your knowledge, your systems, your people.
Knows
The CI/CD runbook: pipeline stages, known failure modes, credential rotation dates, protected gates by engagement and build-time budgets.
Reads from
the source code repository and CI/CD: the pipeline run, its logs, recent commits and the runner and credential status.
Acts in
the source code repository and CI/CD: it records the diagnosis on the run and retries or opens a fix.
Hands back
An object with causes (a ranked list with cause, evidence, confidence and how_to_confirm), fix, retry (true or false), owner and gate_impact (or null).
Asks a person when
The fix changes a protected gate, rotates a credential, or blocks a client release.
Settings
Your limits, set by your managers.
| Setting | Default | What it controls |
|---|---|---|
| Build time budget (minutes) | 15 | Builds slower than this are diagnosed for speed. |
| Flaky test threshold (%) | 5 | A test failing intermittently above this rate is flaky. |
| Automatic retries | 1 | Retries for a failure diagnosed as infrastructure. |
| Protected gates | required tests, security scan, licence scan, client release approval, production deploy approval | Comma-separated gates that are never skipped. |
| Main branch red alert (minutes) | 30 | A main branch failing this long alerts the team. |
Limits are settings, not code: a manager changes them without a developer.
Rules it always follows
- Rank causes by evidence: a change just before the first failure outranks infrastructure, and say what would confirm each.
- Retry up to 1 times only when the cause is infrastructure.
- Never skip or weaken required tests, security scan, licence scan, client release approval, production deploy approval.
- Alert the team when main is red for 30 minutes.
- Diagnose builds slower than 15 minutes and tests flaky above 5 percent.
See the CI/CD Pipeline Agent agent on your own work.
We’ll run it on a sample of your work and show you what your approver would see.