Scans local Git repositories for commits on a given day across all branches, excluding merges. It handles worktrees and rewritten commits, and uses configured repository roots and remote organizations to identify work repositories.
Creates and updates Jira tickets, logs time, and retrieves recently updated tickets. It can assign tickets to the user, set time estimates and time spent, and edit summaries and descriptions.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@jira-work-logger log my work for today"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
An MCP server that turns your git commits into Jira tickets. Point a coding agent such as Claude Code at it and ask "log my work for today": it scans your work repos for your commits on that day, groups them, and creates or updates Jira tickets with estimates and time spent.
scan_daily_work(date): your commits on one day across all work repos, all branches, merges excluded. Worktrees and rewritten commits are listed once.
create_jira_ticket(service_name, title, summary, time_estimate, time_spent, task_type, side): creates a ticket assigned to you, titled [BE] service-name: title (or [FE] / [BE-FE]).
update_jira_ticket(ticket_key, summary, description, time_estimate, time_spent): edits a ticket and logs time.
get_my_tickets(status): your 20 most recently updated tickets in one status.
Related MCP server: jira-mcp
Every failure starts with one line the agent can parse, then says what went wrong and what to do:
jira-work-logger · auth · not retryable
Problem: Jira rejected the API token while trying to create the ticket (401).
Fix: do not retry. Ask the user to create a new API token at id.atlassian.com and put it in config.yaml next to server.py.Categories are config, auth, permission, validation, network and transient, and the first line always says whether a retry makes sense. Durations are checked before anything is sent to Jira, and titles or descriptions that mention AI authorship are refused.
Every git clone directly under git.repo_roots whose origin belongs to one of git.remote_orgs, plus any paths in git.repos. Personal repos in the same folder are skipped.
python3 -m venv .venv
.venv/bin/pip install -r requirements.txt
cp config.example.yaml config.yamlFill in config.yaml:
jira.url, jira.email, jira.api_token (create a token at https://id.atlassian.com/manage-profile/security/api-tokens), jira.project_key
git.author_email: the email on your commits
git.repo_roots and git.remote_orgs: where your work clones live and which GitHub orgs count as work
defaults.issue_type: usually Task
defaults.frontend_services: repo names that get [FE] instead of [BE]
defaults.task_type_field and defaults.task_type_options: optional, for a Jira select field such as "Task type". Example: customfield_10100 with {development: "10200", maintenance: "10201"}
Register it with Claude Code:
claude mcp add jira-work-logger -- /path/to/jira-work-logger/.venv/bin/python /path/to/jira-work-logger/server.pyconfig.yaml is in .gitignore; keep your token out of the repo.
MIT
This server cannot be deployed
Tasks, sprints, releases and docs for engineering teams with AI coding agents.
Assign Linear, Jira or GitHub issues to an AI coding agent team that plans, builds, reviews, tests.
Task management for people and autonomous AI developers: tasks, stories, work logs, pull requests.
Hand tasks, bugs and finished work to AI coding agents, and get back a write-up with evidence.
