Engineering
The request queue moves. So does the roadmap.
Requests that arrive with repro steps, evidence and a definition of done. One agent writes the spec and fixes, another reviews against every criterion, CI passes, and the merge only lands on the branch you chose. By default, nothing reaches production without your team’s decision.
5 deliveries on us, no card needed.
Before
Support writes “the discount is wrong”, you drop what you’re doing, reproduce, fix, open the PR and explain in the channel when it’ll go live.
With Alfrig
The request arrives with repro steps and a definition of done, one agent fixes and another reviews. You check the PR when you want and decide what goes to production.
1Engineering
Connect in minutes
GitHub app (or a token for GitLab, Bitbucket and Azure Repos). Alfrig maps the catalog: stack, test command and dependencies between projects.
2Engineering
Specs with verifiable criteria
Each acceptance criterion says how it’s checked: test, QA or manual. The review checks them one by one.
3Engineering
Review by another agent and CI
The author never approves its own PR. Merge only with green CI, in dependency order, on the agreed branch.
4Engineering
Several repositories in one task
Backend and frontend in the same task, picked from the catalog, with one agent per repository when agents are free and merges in dependency order.
You see everything: spec, diffs, commands, tests, time and cost for every step, and you decide what needs approval.
Control
You decide how much the AI does on its own.
Every step runs on its own or waits for a person. Start by approving every plan and, once you trust it, let everything run through to merge. Try it:
From request to merge with nobody pushing.
1 · Selection
The agent picks the projects from the catalog when the task does not say.
● automatic
2 · Investigation
Reads the repositories and writes the spec: what changes, where and how to verify.
● automatic
3 · Development
Writes the code, runs the tests and opens one PR per repository.
● automatic
4 · Review
Another agent reads the diffs, runs the tests and checks every criterion.
● automatic
5 · Merge
With green CI, squashes in dependency order and deletes the branches.
● automatic
Changes requested in review or a red CI go back for fixing on the same branches, up to a round limit (3 by default).
Where to stop
Fully automatic: the agent picks the projects, investigates and follows its own plan, another agent reviews and, once approved with green CI, merges.You choose what stays automatic at each step.
Also
- QA rejected it? Put the label back and the fix returns to the same branches, from the current code.
- Ask Alfrig to approve, resume, merge, comment on the ticket or change its status: each action becomes a card that only runs once you confirm.
- Approve plans and answer questions in the chat with Alfrig, in a Jira, ClickUp or Linear comment, through a link with no login, or from the buttons on a Slack or Teams alert.
- In Product, approving the proposal and approving the screens can also run on their own or wait for someone.
Real costs
With your key, the model is paid to the provider directly. No markup.
Use your organization’s keys (Anthropic, OpenAI, Google Gemini or z.ai) and pick the model for each step: strong for planning and review, cheaper for building. Calls go through the Alfrig gateway, which uses your key and meters each step, with no markup. Every task shows what it cost. Prefer a single bill? Plans with the Alfrig key include the model.
Tasks
- MergedAdd formatCpf utility and exportPR #392min 16sUS$ 0.02
- MergedThe README doesn’t explain how to run the project on a newcomer’s machinePR #7Linear ↗2min 59sUS$ 0.17
- MergedAdd a local setup section to the READMEPR #372min 12sUS$ 0.14
- Mapping doneMap a 4-project catalog3min 36sUS$ 0.16
- Survey doneSurvey the frontend design system3min 18sUS$ 0.31
- MergedThe project’s AGENTS.mdPR #63min 28sUS$ 0.05
- Read doneRead the project’s code47sUS$ 0.04
Bug fixes in large systems, with deep investigation on the strongest model, came in between US$ 5 and US$ 12. Cost varies with the chosen model and the size of the codebase.
Control and security
Can I trust it with my code? Here are the facts.
Nothing lands without review
Merge only after an approved review and green CI, on the branch you choose: by default, nothing goes straight to production. Auto-merge is optional.
The AI never sees your credentials
Clone, push, PR and merge are done by Alfrig. The agent works on a copy without the repository token.
Isolated, disposable environment
Each step runs in an isolated environment with no inbound connections, destroyed at the end. Tests run offline.
A database of your own
Each organization has its own database. Keys encrypted with AES-256-GCM using a key unique to your organization.
Role-based permissions
Owner, admin, member, requester and guest. Alfrig and connected apps only reach what the person’s role allows.
Deletion that means it
The owner can delete the organization for good: database, files and encryption key. Technical logs kept for up to 2 months, as the privacy policy says.
Integrations
Works where your company already works.
Where requests come from
- Chat with Alfrig
- ClickUp
- Jira
- Linear
- Azure Boards
- API and MCP
Where the code lives
- GitHub
- GitLab
- Bitbucket
- Azure DevOps
Where alerts go
- Slack
- Microsoft Teams
- Desktop app
FAQ
What engineering asks
Is the code any good?+
Every task starts with a spec and acceptance criteria. Tests related to the change run before the PR, and another agent reviews against each criterion, in fix rounds up to a limit (3 by default). If tests fail, it becomes a draft PR, never a merge.
What access does Alfrig need?+
Only to the repositories you choose: through the Alfrig app on GitHub or a token on other providers. Access is encrypted and never handed to the agent.
Does the agent see our credentials?+
No. Clone, push, PR and merge are done by Alfrig; the agent works on a copy with no token, in an isolated environment with no inbound connections, destroyed at the end.
Is our code used to train models?+
Alfrig does not train models. Data use depends on the model provider you connect.
Does it work with our stack?+
Node, PHP, Python, Java/Kotlin, .NET, Go, Ruby and Flutter, plus mixed projects, monorepos included.
Bring in Support and CS
They ask Alfrig directly, and you stop translating requests.
What do you need to fix today?
What do you need?Write to Alfrig. It asks what’s missing and sees it through.5 deliveries on us, no card needed.