Generate Standup Notes from Commit History

Standup Notes Generator is a skill that positions the model as a team communication specialist focused on async-first standup practices, AI-assisted note generation from commit history and remote team coordination. It generates daily standup notes by analyzing several data sources: Git commit history, Jira tickets, an Obsidian vault and calendar events. It supports both traditional synchronous standups and async-first team communication, extracting accomplishments from commits and formatting them for team visibility.

If you pass arguments, they are used as context about specific work areas, projects or tickets to highlight. If you pass none, the skill discovers work automatically from all available sources. It expects the mcp-obsidian integration for vault access and the atlassian integration for Jira queries (with a graceful fallback when unavailable), and calendar integrations are optional.

The collection strategy has five steps: get the current user context (Jira username and Git author); fetch recent commits and parse them for PR references, ticket IDs and features; query Obsidian for recent changes and periodic notes; search Jira for completed, in-progress and planned high-priority tickets; and correlate everything, linking commits to tickets and tickets to notes. The Git step uses a command like this:

git log --author="<user>" --since="yesterday" --pretty=format:"%h - %s (%cr)"

The standard note has three sections: Yesterday or Last Update, Today or Next, and Blockers or Notes, with an optional links footer. Formatting guidelines are to use bullets for scanability, link tickets, PRs and docs, bold blockers, add time estimates or completion targets, keep each bullet to one or two lines, and group related items.

The skill spells out how each section is built:

• Accomplishments come from commits (conventional commit types, ticket references and the described action, grouped by feature area, ticket or type of work) and from completed Obsidian tasks. Good accomplishments focus on delivered value rather than activity and include impact when known.
• Plans are prioritized: urgent blockers for others first, then sprint commitments, high-priority bugs or production issues, work in progress, code reviews and support, and finally new backlog work. Planning is capacity-aware (8 hours minus meetings and expected interruptions) and each task gets a clear outcome.
• Blockers are categorized as hard (work completely stopped) or soft (work slowed but not stopped), with an escalation format that gives the impact, the specific need, who it is from, what was tried and the next step if unresolved by a given date. Cross-team dependencies are called out with expected delivery dates and @mentions.
• AI summarization groups related commits under one bullet, translates commit messages into business value, spots patterns and flags likely blockers. A manual override step tells you to review generated content for accuracy and add context the AI cannot infer.

The playbook also covers communication practices, including visibility and transparency (share wins as well as problems, be honest early about timeline concerns, call out dependencies before they become blockers), team coordination (read teammates' notes before posting, offer help on blockers you can resolve, use threads and keep the main post scannable) and writing style (active voice, specific timelines, a casual human tone). Its follow-up tracking flags any item that stays in the Today section for 3 or more days as potentially stuck work. The async-first principles add posting at a consistent time, including context for readers in other timezones, making blockers actionable and avoiding ambiguous words like "soon". The playbook describes several async formats (written-only in a Slack channel, thread-based, a 2-3 minute Loom video with a text summary, and a rolling 24-hour standup), follow-up tracking that turns blockers and promises into tasks, and a monthly retrospective on estimate accuracy and common blockers.

Two full reference examples close the playbook. The first is a distributed team across US, Europe and Asia timezones with no synchronous standup: a 30 minute morning routine that drafts the note from Git, Jira and yesterday's Obsidian daily note, reviews teammates' notes from the last 8 hours, posts to a Slack standup channel at 9:00 AM local time, checks thread responses by 11am, and updates the task list, followed by a Friday retrospective that reviews patterns such as completed stories, blockers that cost time and code review throughput. The second is an AI-powered generation system in four layers: data collection (Git commits, Jira updates, Obsidian changes, calendar events and Slack activity), AI analysis and correlation (linking commits to tickets, grouping related commits, extracting business value, spotting blockers from patterns such as repeated attempts, and calculating work distribution between features, bugs and reviews), generation and formatting for Slack, Discord, email or Obsidian, and a human review layer where the person adds context the AI cannot infer, adjusts priorities and approves the post. Thread-based standups use emoji reactions as quick acknowledgments, one for read, one for noted and one for offering help.

Use it when a remote or distributed team wants daily standup notes generated from real work data rather than written from memory, or wants to move to written, async updates without meetings.

Do not use it when the task is unrelated to standup notes or needs a different domain or tool outside this scope. The skill also states that its output is not a substitute for environment-specific validation, testing or expert review, and that you should stop and ask if required inputs, permissions, safety boundaries or success criteria are missing.

The inputs are optional context about specific projects or tickets, plus access to Git history, Jira, an Obsidian vault and optionally a calendar. The output is a structured standup note with yesterday, today and blockers sections, formatted for the target platform (Slack, Discord, email or Obsidian), and optionally a list of follow-up tasks extracted from the note.

The skill names mcp-obsidian, the atlassian Jira integration, Git, calendar integrations, Slack or Discord for posting, and Loom for video standups. Its extraction commands include Obsidian recent changes and periodic notes, and Jira queries on tickets that changed to Done in the last day or are In Progress.

Engineers and team leads on remote or distributed teams who post daily standup updates and want them drafted from commits, tickets and notes.