project-sync

project-sync

On-demand AI-powered sync of a project's repo state into the brain vault. Reads live git/GitHub/Jira data, updates the project status doc, and makes targeted edits to relevant brain docs where content is clearly stale.

Trigger phrases

• /project-sync [project-name]
• "sync acme"
• "update project docs for [project]"
• "what's the current state of [project]"

Default project: set your own default, or require an explicit argument.

Config

Load project config from ~/brain/scripts/repos.json. Each entry has:

• repo_path — local repo path
• gh_repo — GitHub owner/repo for gh CLI
• jira — (optional) { "project_key": "ACME", "site": "your-org.atlassian.net" }
• status_doc — AI-owned status file (always overwritten)
• brain_docs — list of brain notes to review and potentially update

Step-by-step workflow

1. Gather repo data

Run these commands against the project's repo_path:

# Recent commits (last 14 days)
git -C {repo_path} log --oneline --since="14 days ago" --format="%h %s (%an, %ar)"

# Active branches (excluding stale)
git -C {repo_path} branch -r --sort=-committerdate | head -20

# Current branch + status
git -C {repo_path} status --short

# Open PRs
gh pr list --repo {gh_repo} --json number,title,author,labels,createdAt,url --limit 20

# Recently merged PRs (last 7 days)
gh pr list --repo {gh_repo} --state merged --json number,title,mergedAt,url --limit 10

2. Gather Jira data (if configured)

If the project has a jira config, use acli to pull issue data:

# Recently updated issues (last 14 days)
acli jira workitem search --jql "project = {project_key} AND updated >= -14d ORDER BY updated DESC" --fields "key,summary,status,assignee,priority" --limit 30 --json

# Current sprint issues
acli jira workitem search --jql "project = {project_key} AND sprint in openSprints() ORDER BY priority ASC" --fields "key,summary,status,assignee,priority" --limit 50 --json

# Blocked / flagged issues
acli jira workitem search --jql "project = {project_key} AND status = Blocked ORDER BY priority ASC" --fields "key,summary,status,assignee,priority" --json

# Issues due this week
acli jira workitem search --jql "project = {project_key} AND duedate <= endOfWeek() AND status != Done ORDER BY duedate ASC" --fields "key,summary,status,assignee,duedate" --json