Integration · Developer tools

Connect Claude & ChatGPT to Bitbucket

Ask your AI which pull requests are waiting, what a change does, or why the last pipeline failed in Bitbucket. It reads repositories, branches, commits, files and diffs, and can open, comment on, approve and merge pull requests. It also reads pipeline runs and step logs, so you can follow up on code and builds from the chat.

Connect Bitbucket free

Free plan, no credit card. Takes about a minute.

Try asking

“Which pull requests are open in the api repository?”
The AI lists the open pull requests with title, author, source and destination branch.
“Review pull request 57 and point out anything risky.”
It reads the diff and the changed files with lines added and removed, then writes a short review with the parts worth a closer look.
“Why did the last pipeline on main fail?”
It finds the newest pipeline run, lists its steps to see which one failed and reads that step's log.

What your AI can do

Pull requests

List, open and update pull requests, read diffs, files, commits and comments, and comment, approve, request changes, decline or merge.

Code and history

List repositories, projects and branches, read files and folders at any branch or commit, and look up commits.

Pipelines

List pipeline runs, see which step failed and read the step log.

“Open a pull request from feature/login-fix to main and add Sara as reviewer.”
It looks up Sara's account ID among the workspace members and opens the pull request with title, summary and reviewer.
“Approve PR 57 and squash-merge it.”
It approves the pull request as you, then merges it with the squash strategy.
“Show me config/settings.yml on the release branch.”
It reads the file's contents at the head of that branch.

27 tools for Bitbucket

These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.

approve_pull_requestVisible to othersApprove a pull request as the signed-in user.
comment_on_pull_requestChanges dataAdd a comment to a pull request. Put the text in `content` ({"raw": "..."}); add `inline` with a file path and line to comment on code, or `parent` to reply.
create_pull_requestChanges dataOpen a pull request with `title`, `source` branch, optional `destination` branch (defaults to main), `summary` and `reviewers`. Set `type` to 'pullrequest'.
decline_pull_requestVisible to othersDecline (close without merging) a pull request. A declined pull request cannot be reopened.
get_commitGet one commit by hash, with author, date, message and parent commits.
get_current_userGet the signed-in user's display name, account ID and UUID.
get_file_contentsRead a file's contents, or list a folder, at a `commit` hash or branch name and a `path` in the repository.
get_pipelineGet one pipeline run by `pipeline_uuid`, with its state, result, branch, commit and duration.
get_pipeline_step_logGet the log output of one pipeline step by `pipeline_uuid` and `step_uuid`.
get_pull_requestGet one pull request by `pull_request_id`: title, description, source and destination branches, reviewers, approvals and state.
get_pull_request_diffGet the full code diff of a pull request as unified diff text.
get_repositoryGet one repository by `repo_slug`, with its description, main branch, language and last update.
list_branchesList the branches in a repository, optionally filtered with `q` such as 'name ~ "feature"'.
list_commitsList recent commits in a repository, newest first.
list_pipeline_stepsList the steps of a pipeline run with each step's status, to see which step failed.
list_pipelinesList pipeline (CI/CD) runs for a repository, with status, branch and trigger; sort with `sort` '-created_on' for newest first.
list_projectsList the projects that group repositories in your workspace.
list_pull_request_commentsList the comments on a pull request, including inline code comments.
list_pull_request_commitsList the commits included in a pull request.
list_pull_request_filesList the files changed in a pull request, with lines added and removed per file.
list_pull_requestsList pull requests in a repository; filter by `state` OPEN, MERGED, DECLINED or SUPERSEDED (open by default).
list_repositoriesList the repositories in your workspace, optionally filtered with a `q` query such as 'name ~ "web"' and sorted with `sort` (e.g. '-updated_on').
list_user_pull_requestsList the pull requests authored by one person (`selected_user`, their account ID or UUID) across the workspace, filtered by `state`.
list_workspace_membersList the people who are members of your workspace, with their account IDs for reviewers.
merge_pull_requestVisible to othersMerge a pull request into its destination branch, with optional `message`, `merge_strategy` (merge_commit, squash, fast_forward) and `close_source_branch`. Set `type` to 'pullrequest'. The merge cannot be undone from here.
request_changesChanges dataMark a pull request as needing changes as the signed-in user.
update_pull_requestChanges dataChange an open pull request's title, description, destination branch or reviewers. Set `type` to 'pullrequest'.

Set up in three steps

1
Pick the app
Create a free PipMCP account and choose Bitbucket from the app list.
2
Paste your key
Enter your Atlassian account email, a Bitbucket API token and your workspace ID (the short name in bitbucket.org/<workspace>/). To create the token, open Personal settings > Atlassian account settings > Security > Create and manage API tokens, choose Create API token with scopes, pick Bitbucket and the permissions you need.
3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
Click your name, then Settings › Connectors › Add custom connector. Paste your link as the Remote MCP server URL. Switch it on from the + menu in a chat.

Questions

What can the AI do in Bitbucket?

The AI can read repositories, projects, branches, commits, files, pull requests, diffs, comments, pipeline runs and step logs. It can open and update pull requests, comment on them, approve or request changes, decline them and merge them. It only reaches what your API token and its scopes allow.

Does the AI see my Bitbucket API token?

No. Your Bitbucket credentials are encrypted at rest and never shown to the AI. After you save them, they are not shown again, not even to you. The AI only sees the results of the tools it calls.

Can I control what the AI is allowed to do?

Yes. You choose which tools are switched on, so you can start read-only. Merging and declining a pull request cannot be undone from here, and both can require your approval before they run. Every tool call is logged.

Does it work with ChatGPT?

Yes. In ChatGPT go to Settings › Apps & Connectors › Advanced and turn on Developer mode, then add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.

What does it cost?

PipMCP has a free plan with no credit card required. Paid plans bill per completed task. You also need your own Bitbucket account.

Why do I enter a workspace ID?

Almost every Bitbucket request is made inside a workspace. PipMCP stores your workspace ID with the connection, so the AI never has to ask for it and only works in that workspace. PipMCP is an independent service and is not affiliated with Atlassian.