“Who approved that release?”
It shipped on a Friday night. Nobody can say if anyone checked it first.
engineering: Ship with the process attached.
ConvOps runs your engineering process as governed AI agent workflows. Bugs reproduce before they get fixed. Tests go red before green. Releases wait for your OK. Every engineer and agent runs it the same way.
claude code · cursor · codex · gates the engine checks · nothing to install
new chat
task
step 1 of 6EU cards fail at checkout
bug arrives reproduce reproduced? test + fix approve approvalrelease
the problem
Fixed. But how, and by whom?
Every engineer and every agent fixes bugs a little differently. The reasons vanish with the chat.
“Where did that decision go?”
Why we pinned the payments library lives in a closed chat. The next fix undoes it.
“What happens when someone leaves?”
Their way of debugging leaves with them. The new hire and the agent start from zero.
watch it work
One bug, report to release.
A real bug, run through the Bug Fix workflow from Claude Code. Same steps, whoever picks it up.
01you, in Claude Code
The bug becomes a task.
You paste the error and ask for a fix. ConvOps files a bug and attaches the Bug Fix workflow. The work has a home before anyone touches code.
example
Checkout fails for EU cards. Error log attached.
bug filed · Bug Fix workflow attached
Step 1 of 6: reproduce. Starting there, before any fix.
02the AI
It looks before it touches code.
Before acting, the AI asks the brain: past decisions on payments, where checkout lives, related bugs. It starts where the last person stopped.
example
askeu card checkout failure
memoryEU cards need the 3-D Secure step. Never skip it.
routeCheckout code: api/payments/checkout/
taskCard declines in Spain (done)
03the AI · reproduce
No fix before the failure is real.
The AI writes a test that fails the way production fails. Cannot reproduce it? The bug closes with an explanation instead of rotting.
example
- $ pytest tests/checkout -k card
- ✓ test_us_card_checkout
- ✓ test_uk_card_checkout
- ✕ test_eu_card_checkout
- 3-D Secure redirect lost the session
reproducedFailing test captured on the task
04the AI · fix
Red, then green.
The fix lands against the failing test. The test turns green, and the whole checkout suite runs again before anyone is asked.
example
- - cookie.same_site = "Strict"
- + cookie.same_site = "Lax"
- $ pytest tests/checkout
- ✓ 3 passed
fix readyEvery checkout test green
05you approve
The release waits for you.
The workflow stops at the approval gate and will not advance without it. You read the fix, the test and the reasoning, then say go.
example
waiting for you
Release the EU card fix?
1 file changed · 3 tests green · reproduced first
Send backApprove release
approvedRelease step unlockedThe run continues from exactly here.
06the record
Logged. And remembered.
Every step lands on the task, with the agent and the signed-in person kept apart. The lesson becomes a memory, so the next bug starts smarter.
example
- createdBug filed from Claude Code
- step doneReproduce: failing test captured
- step doneFix: 3 tests green
- approvedRelease gate passed
- completedReleased
agent: claude-code (its own label) · person: from verified sign-in
memory · lesson3-D Secure redirects drop strict cookies. Use Lax.linked to this task
example
Checkout fails for EU cards. Error log attached.
bug filed · Bug Fix workflow attached
Step 1 of 6: reproduce. Starting there, before any fix.
the workflow
The Bug Fix workflow, running.
Six steps and one approval. The same for every engineer, every agent and every bug.
workflow · Bug Fixrunning · step 1 of 6
- 01
A bug arrives
Typed in chat, sent from your stack, or found by another workflow.
- 02
Reproduce
No fix before the failure is real and captured.
- 03
Branch on reality
Cannot reproduce? It closes with an explanation instead of rotting.
decision - 04
Failing test, then the fix
Red before green. The fix ships with its proof.
- 05
You approve the release
The run holds here until a person says go.
needs a person - 06
Release
Logged on the task, step by step.
what it remembers
Every bug teaches the next.
Kept by the team, not by whoever was on call. Recalled before the next fix starts.
Decisions
Why the payments library is pinned, with the reason and the task that decided it.
why we did it
Where code lives
Checkout lives in api/payments/checkout. No session searches for it again.
where it lives
What broke before
The fix that broke Safari last spring, so nobody ships it twice.
lessons from failures
Linked to the bug
Every memory points back to the task that produced it. The reason and the code stay together.
traceable
the difference
Ad hoc vs taught once.
- Each engineer fixes bugs their own way
- The agent jumps straight to a fix
- Releases ship when someone feels ready
- The reason for a change lives in a closed chat
The gate is real
Approval is a requires_approval gate. The engine refuses to advance the cursor until it is met, and returns what is missing.
Routing attaches the workflow
A workflow assignment maps task type, tags or project to a template, so a bug gets Bug Fix the moment it is filed.
Detours come back
A decision step records its outcome. Detours, like closing a bug that will not reproduce, run as sub-workflows that return to the main line.
Runs in your client
Attended runs need only the MCP connection. For unattended work, a custom runner on your machines receives an envelope and a lease.
start this week
Four moves to your first run.
Connect the AI tool your team already uses, install the workflows, and run one real piece of work through them.
- move 1
Connect your client and install Bug Fix and Feature from the catalog
- move 2
File one real bug. Watch the workflow attach itself and hand your agent step one
- move 3
Sit at the approval gate: the release waits until you say go
- move 4
Edit one step instruction. The next run inherits it, and nobody reads a memo
install these
engineering
Bug FixResolve bugs from report to verified fix. The AI drives through Intake, Reproduce, Analyze, Fix, and Verify.engineering
FeatureDeliver features from requirements to reviewed release. The AI drives through Capture, planning, execution, and Verify.universal
PlanTurn strategic work into reviewed, executable plans. The AI drives through Analyze, Draft, phases, and execution.universal
PhaseExecute an implementation phase from ready to verified. The AI drives through Ready, Implement, and Verify.universal
TaskTake any work item from intake to verified completion. The AI drives through Intake, Analyze, Implement, and Verify.built on
The parts doing the work.
Workflow engineSteps and gates the AI follows.GovernanceApprovals the run cannot skip.The brainDecisions and lessons, recalled.ConnectOne line in Claude Code or Cursor.
ConvOps is the operations layer for AI agents: an MCP server that holds your team's process as workflows, with approval gates, a shared memory and an audit trail. It runs no AI models.
questions
What teams ask.
Does this replace our issue tracker?
It becomes the system of record for AI-driven work: tasks with hierarchy, workflows, and a full query surface your agents reach over MCP. Whether it replaces or sits beside your tracker is your call; nothing forces a migration to start.
Which coding clients work?
Claude Code, Cursor, ChatGPT, and any client that speaks MCP. The engine hands out steps; whichever client is connected does the work.
Can an agent skip the process when nobody is looking?
Routing attaches the workflow at task creation, and gates are evaluated by the engine on every advance. A run that has not satisfied a gate does not move, whoever or whatever is driving.
Can implementation run on our own machines?
Yes. Attended runs already execute in your local client. For unattended work, a custom runner on your infra receives an envelope and a lease, or you point dispatch at your own OpenCode server.