> ## Documentation Index > Fetch the complete documentation index at: https://docs.ankra.ai/llms.txt > Use this file to discover all available pages before exploring further. # AI Gateway > The AI Gateway connects Ankra's AI to Slack, Microsoft Teams, GitHub, GitLab and Bitbucket - how a message or a pull request event reaches the AI and back, Ask and Agent safety modes per connection, what each connection may do, the staging cluster for workspaces and previews, and the settings that govern it. The AI Gateway is how Ankra's AI meets your team where it already works. A question in Slack, an `@ankraai` in a pull request, a push that needs a review, a failed build that needs a diagnosis: each arrives through one gateway, is answered by the same AI with the same knowledge of your clusters and pipelines, and goes back to where it came from. ```mermaid theme={null} flowchart LR subgraph Chat["Chat"] Slack[Slack] Teams[Microsoft Teams] end subgraph SCM["Source control"] GH[GitHub] GL[GitLab] BB[Bitbucket Cloud] end Slack --> GW Teams --> GW GH --> GW GL --> GW BB --> GW GW[AI Gateway
one profile per connection:
mode + capabilities] --> AI[Ankra AI
your clusters, pipelines,
runs, findings, docs] AI --> Lanes{What was asked} Lanes --> Answer[Answer in the thread] Lanes --> Review[Pull request review] Lanes --> Fix[Diagnosis or fix pull request] Lanes --> Preview[Preview or workspace
on the staging cluster] ``` ## Every connection has a profile Each connection (a Slack workspace, a Teams tenant, a GitHub App installation, a GitLab token, a Bitbucket workspace) carries a **profile**: a safety mode, and the capabilities switched on for it. | Connection | Starts with | Capabilities you switch on | | - | - | - | | Slack, Microsoft Teams | Chat, alert actions and notifications. Anything that changes something asks first | Follows the organisation's default mode | | GitHub, GitLab, Bitbucket Cloud | **Everything off.** Installing an app or connecting a token grants access, nothing more | AI review, review drafts, mention replies, PR previews, push code review, failed-build review, skills review, merge when green | Source-control connections start off because a pull request is a shared, public place. Nothing should post there until someone decides it should. ## Ask and Agent The mode bounds **how far** an answer may go. The capability switches decide **whether** the AI answers at all. ```mermaid theme={null} flowchart LR subgraph Ask["Ask (default)"] A1[Investigate and explain] A2[Read runs, logs, findings] A3[Safe creations:
workspace pod, PR demo,
a brand-new Stack] end subgraph Agent["Agent: Can act"] G1[Everything in Ask] G2[Open fix pull requests] G3[Merge when green] G4[Other changes
still ask to confirm] end ``` | | Ask | Agent | | - | - | - | | Explain a failing pipeline run | Yes | Yes | | Spin up a throwaway workspace or PR demo | Yes | Yes | | Change existing infrastructure | No | With confirmation | | Open a fix pull request | No | Yes, and the pull request is the review gate | | Merge when green | Refused, naming the setting | Yes | * **Ask is the default**, and the fallback for a connection with no mode set. * **Automatic events stay read-only in either mode.** A review comments. A proposed fix arrives as a separate pull request you review like any other. * **An unconfigured chat connection** inherits the organisation's default mode: ```bash theme={null} ankra org ai-environment set --default-gateway-mode ask ``` The same two modes exist everywhere: `ankra chat --mode ask|agent` in the CLI, and `mcp:read` and `mcp:write` tokens on the [MCP server](https://docs.ankra.ai/pipelines/cli-and-mcp.md). ## From a Slack question to an answer ```mermaid theme={null} sequenceDiagram autonumber actor You participant Slack participant GW as AI Gateway participant AI as Ankra AI participant P as Ankra Pipelines You->>Slack: @Ankra why is main red? Slack->>GW: Event GW->>GW: Is this Slack user linked to an Ankra account? alt not linked yet GW-->>Slack: Connect your Ankra account You->>GW: Link (once) GW->>GW: Replay the parked question end GW->>AI: Ask, as you, with your permissions AI->>P: Latest run on main, failing step, log tail P-->>AI: e2e failed, exit 1, findings AI-->>Slack: Streamed answer in the thread You->>Slack: Reply in the thread Slack->>GW: Continues the same conversation ``` * **The AI answers as you.** Each person links their Slack or Teams account to their Ankra account once. From then on the AI has exactly that person's permissions, never more. * **Threads are conversations.** Mention the bot to start one, and reply in the thread to continue it. * **Ticket threads sync with the board.** A reply that starts with `Decision:` answers a pending choice on a board ticket. * **`/ankra status`, `/ankra logout` and `/ankra help`** work in Slack. Prompts that work from Slack or Teams: > @Ankra is the pipeline for my-org/my-repo green on main, and if not, what broke? > @Ankra show me the failing step's log for the last pipeline run on pull request 412 in my-org/my-repo. > @Ankra the nightly e2e failed again. Is it the same test as yesterday? ## From a pull request event to a review Provider webhooks feed separate lanes, each with its own switch. Every event is deduplicated, so a redelivered webhook never reviews twice. ```mermaid theme={null} flowchart LR Event[Webhook: opened, pushed,
ready, comment, closed] --> Dedup[Deduplicate] Dedup --> R[AI review lane] Dedup --> M[Mention lane: @ankraai] Dedup --> Pv[Preview lane] Dedup --> Ac[Failed build lane] Dedup --> Am[Merge when green lane] R & M & Pv & Ac & Am --> SC[One status comment
plus inline findings] ``` What each lane does is on [AI in pull requests](https://docs.ankra.ai/pipelines/ai-in-pull-requests.md). What each switch means is on [AI Code Review](https://docs.ankra.ai/platform/ai-code-review.md). ## The staging cluster Workspaces, pull request previews and PR demos run on one cluster your organisation nominates for AI work: the **staging cluster**. If your organisation has no separate CI cluster, Ankra Pipelines runs there too. ```mermaid theme={null} flowchart LR AI[Ankra AI] --> WS[Workspace pod
locked down, shallow clone,
time-limited] PR[Pull request] --> PV[Preview
own namespace,
public https URL] WS --> Staging[(Staging cluster)] PV --> Staging DNS["*.previews.example.com
your wildcard DNS"] --> PV ``` ```bash theme={null} ankra org ai-environment get ankra org ai-environment set --staging-cluster staging ankra org ai-environment set --workspace-ttl-hours 12 --demo-ttl-hours 48 ankra org ai-environment set --demo-ttl-role-cap member=24 --demo-ttl-role-cap operator=96 ankra org ai-environment set --demo-base-domain previews.example.com ``` * **Workspaces** are locked-down, time-limited pods that shallow-clone a repository so the AI can read and test code. `--workspace-ttl-hours` takes 1 to 168. * **Previews and demos** run in their own namespace with a public https URL. `--demo-ttl-hours` takes 1 to 720, and `--demo-ttl-role-cap` caps it per role. * **The preview domain is yours.** Ankra publishes no DNS for `--demo-base-domain`: point a wildcard record at the staging cluster's ingress. The command tells you when a setting would leave previews on plain http. ## Who can change what | Action | Who | | - | - | | Connect Slack, Teams, GitHub, GitLab or Bitbucket | Organisation administrator | | Change a connection's mode or capabilities, per connection or per repository | Organisation administrator | | Set the staging cluster and its time limits | Organisation administrator | | Ask the AI in chat | Any linked member, with their own permissions | | Request merge when green | A member with `pipelines.operate` and a verified GitHub identity, on an Agent-mode connection | | Turn an application's auto-fix off | Deploy permission on that application | ## Related * [AI connections](https://docs.ankra.ai/platform/ai-connections.md): connecting Slack and Teams step by step. * [GitHub](https://docs.ankra.ai/integrations/github.md), [GitLab](https://docs.ankra.ai/integrations/gitlab.md) and [Bitbucket Cloud](https://docs.ankra.ai/integrations/bitbucket-cloud.md): the source-control integrations and their permissions. * [AI workspaces](https://docs.ankra.ai/platform/ai-workspaces.md): what a workspace can and cannot do. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.