Daily engineering standup digest in Slack

Build me an agent workflow that posts a daily engineering standup digest in Slack every weekday at 8:30am. Trigger it on a cron schedule (Mon to Fri, 8:30am in our team's timezone).

Inputs I will configure: the Jira board ID for our active sprint, the list of GitHub repos owned by the team (owner/name pairs), the destination Slack channel, and a small mapping of team members from Jira display name or accountId to GitHub username.

Step 1 — Pull the sprint from Jira. Use Get Board Sprints with state=active to find the current sprint for the configured board, then use Get Sprint Issues to load every issue in that sprint. For each issue grab the summary, status, story points, assignee, and the issue key. For any issue that looks at-risk (status not Done and updated recently, or with words like 'blocked', 'blocker', 'waiting on' in a recent comment), call Get Comments on that issue and pull the last one or two comments so we can quote a real blocker. Group the resulting issues by assignee.

Step 2 — Pull pull requests from GitHub. For each configured repo, use List Pull Requests to fetch open PRs (state=open) and recently merged PRs (state=closed, sorted by updated desc, filtered to those merged in the last 24 hours). For each PR keep the title, number, URL, author login, draft flag, mergeable state, requested reviewers, and merged_at. Then bucket PRs by author login.

Step 3 — Join Jira assignees with GitHub authors using the team mapping I provide so each engineer has a single bundle of {sprint issues, open PRs, merged PRs, blocker comments}.

Step 4 — For each person with any activity, have the agent write a short standup in this shape: a one-line 'Yesterday' (merged PRs and Jira issues moved to Done), a one-line 'Today' (in-progress sprint issues and open PRs), and a 'Blockers' line only when there is a real signal (a blocker comment, a PR stuck waiting on review, an issue with no movement). Keep each person to roughly three to five lines. Skip anyone with zero sprint issues and zero PR activity rather than padding the report with 'no updates'.

Step 5 — Assemble one Slack message: a header like 'Daily standup, {date}' followed by a section per person. Use Slack mrkdwn formatting, link Jira issue keys to their Atlassian URLs and PR numbers to their GitHub URLs. Post it to the configured channel using Send a Message in Slack. If the active sprint or both data sources come back empty, post a single short note like 'No sprint activity to report today' instead of a long empty digest.

Be defensive about partial failures: if GitHub fails but Jira works, still post the Jira-only digest with a note that PR data was unavailable, and vice versa. Do not retry the whole run; just degrade gracefully.