summary-init
Skill0 starsChanged 7 months agoMITView on GitHub
Interactive setup wizard for work-summary-kit. Walks you through configuring your data sources, paths, and preferences. Generates ~/.config/work-summary/config.json. Run /summary-init to set up or reconfigure your work summary system.
In this repository3 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- daily-summarySkill
- weekly-summarySkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
23 rules. About 0 are common, 19 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈19
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 21
Ask which data sources you use
- Distinctive: line 23
Verify prerequisites (env vars, MCP servers)
- Distinctive: line 26
Run a quick validation
- Distinctive: line 34
If a config already exists, ask the user:
- Distinctive: line 37
If reconfiguring, read the existing config as defaults for the questions below.
- Distinctive: line 41
Ask the user for:
- Distinctive: line 53
Ask the user which data sources they want to enable.
- Very specific: line 77
Verify: cd {first_repo} && git log --author="{email}" -1 --oneline to confirm commits exist
- Distinctive: line 92
If not set, tell the user:
- Distinctive: line 100
Use defaults for status mappings (done, in-progress, handed-off).
- Very specific: line 110
Use defaults: calendar_id: "primary", working hours 9-6, standard meeting categories.
- Distinctive: line 130
Ask about channel categories:
- Distinctive: line 152
JIRA: Ask for assignee email (default to JIRA email).
- Distinctive: line 154
Obsidian: Ask for the absolute path to the Task Notes directory.
- Distinctive: line 161
Google Tasks: Ask for list ID or use default (@default).
- Very specific: line 163
Linear: Ask for team key. Check LINEAR_API_TOKEN env var.
- Distinctive: line 175
Use defaults for file naming and sections order.
- Distinctive: line 182
If yes, help them define a few mappings.
- Distinctive: line 187
If yes, help them define a few confirmed mappings.
- Very specific: line 197
Write the config to ~/.config/work-summary/config.json.
- Distinctive: line 218
Config readable: Read back the config and confirm it parses
- Distinctive: line 220
Env vars set: Check all required env vars are non-empty
- Distinctive: line 241
If yes, invoke the daily-summary skill.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Work Summary Kit Setup Wizard
Overview
This wizard creates your personal config at ~/.config/work-summary/config.json. It will:
- Gather your basic info (name, email, timezone)
- Ask which data sources you use
- Configure each enabled source
- Verify prerequisites (env vars, MCP servers)
- Generate the config file
- Create output directories
- Run a quick validation
Step 1: Check for Existing Config
test -f ~/.config/work-summary/config.json && echo "exists" || echo "none"If a config already exists, ask the user:
You already have a config at
~/.config/work-summary/config.json. Do you want to reconfigure from scratch or update specific settings?
If reconfiguring, read the existing config as defaults for the questions below.
Step 2: Basic Info
Ask the user for:
- Name: "What name should appear in your summaries?"
- Email: "What's your primary work email?" (used for git filtering, calendar queries)
- Timezone: "What's your timezone?" Detect system timezone as default:
Offer detected timezone as default, let user override.readlink /etc/localtime 2>/dev/null | sed 's|.*/zoneinfo/||' || echo "America/Chicago"
Step 3: Data Sources
Ask the user which data sources they want to enable. Present as a multi-select:
Which data sources do you want to include in your summaries?
- Git -- Commit activity from your repos (recommended for everyone)
- JIRA -- Ticket status and progress tracking
- Google Calendar -- Meeting tracking and focus time analysis
- Slack -- Message activity and key discussions
- Google Drive -- File modification tracking
- Task Management -- Track task completions (Obsidian, JIRA, Google Tasks, or Linear)
Step 4: Configure Each Enabled Source
Git Configuration
Ask:
- "What git repos should I track? Provide the absolute path to each."
- Help: "You can add multiple repos. I'll scan them for your commits."
- For each repo, detect the name from the directory and default branch:
basename {path} cd {path} && git symbolic-ref refs/remotes/origin/HEAD 2>/dev/null | sed 's|refs/remotes/origin/||'
- "What email do you use for git commits?" (default to the email from Step 2)
- Verify:
cd {first_repo} && git log --author="{email}" -1 --onelineto confirm commits exist
- Verify:
JIRA Configuration
Ask:
- "What's your JIRA instance URL?" (e.g., https://company.atlassian.net)
- "Which JIRA project keys should I track?" (e.g., ENG, PLAT, FE)
- "What email do you use for JIRA?" (default to primary email)
Check env vars:
test -n "$JIRA_USERNAME" && echo "JIRA_USERNAME is set" || echo "WARNING: JIRA_USERNAME not set"
test -n "$JIRA_API_TOKEN" && echo "JIRA_API_TOKEN is set" || echo "WARNING: JIRA_API_TOKEN not set"If not set, tell the user:
Set these in your shell profile (~/.zshrc or ~/.bashrc):
export JIRA_USERNAME="your-email@company.com" export JIRA_API_TOKEN="your-api-token"Get your API token at: https://id.atlassian.com/manage-profile/security/api-tokens
Use defaults for status mappings (done, in-progress, handed-off). Mention the user can customize these later.
Google Calendar Configuration
Requires the Google Workspace MCP server. Check if it's available:
Tell the user:
Calendar integration requires the Google Workspace MCP server for Claude Code. Make sure it's configured in your Claude Code settings.
Use defaults: calendar_id: "primary", working hours 9-6, standard meeting categories.
Slack Configuration
Ask:
- "What's your Slack username?"
Check env var:
test -n "$SLACK_TOKEN" && echo "SLACK_TOKEN is set" || echo "WARNING: SLACK_TOKEN not set"If not set, tell the user:
Set your Slack user token in your shell profile:
export SLACK_TOKEN="xoxp-your-token-here"You need a user token (xoxp) with
search:readscope. Create one at: https://api.slack.com/apps (create an app, add OAuth scopes, install to workspace)
Ask about channel categories:
Want to set up channel categories? (Optional -- helps group Slack activity in summaries)
- Project channels (e.g., eng-, proj-)
- Coordination channels (e.g., general, engineering)
- Support channels (e.g., ask-, help-)
Google Drive Configuration
Requires the Google Workspace MCP server (same as Calendar). No additional config needed.
Task Management Configuration
Ask:
Which task management system do you use?
- None -- Skip task tracking
- JIRA -- Use JIRA tickets as your task system (uses same JIRA config)
- Obsidian -- Obsidian Task Notes (markdown files with YAML frontmatter)
- Google Tasks -- Google Tasks lists
- Linear -- Linear issues
Based on selection:
JIRA: Ask for assignee email (default to JIRA email). Uses same auth as JIRA data source.
Obsidian: Ask for the absolute path to the Task Notes directory. Explain the expected format:
Obsidian Task Notes should be markdown files with YAML frontmatter containing:
status: open, in-progress, blocked, donecompletedDate: YYYY-MM-DD (set when done)project: project namepriority: priority level
Google Tasks: Ask for list ID or use default (@default). Requires Google Workspace MCP.
Linear: Ask for team key. Check LINEAR_API_TOKEN env var.
Step 5: Output Preferences
Ask:
- "Where should summaries be saved?" (default: ~/work-summaries)
- "Which template do you prefer?"
- default -- Full detailed summary with all sections
- compact -- Quick snapshot: summary, metrics, blockers only
- engineering -- Heavy on code, light on meetings (coming soon)
- manager -- Heavy on meetings/decisions (coming soon)
Use defaults for file naming and sections order. Mention these are customizable in the config.
Step 6: Optional Settings
Ask:
Do you want to set up project name mappings? (Maps folder names to friendly display names)
If yes, help them define a few mappings.
Ask:
Do you want to set up name mappings? (Override display names for colleagues)
If yes, help them define a few confirmed mappings.
Step 7: Generate Config
Assemble the config JSON from all collected answers. Use defaults for anything not explicitly configured.
mkdir -p ~/.config/work-summaryWrite the config to ~/.config/work-summary/config.json.
Also detect and set kit_path:
# Try to find the kit installation
SKILL_DIR=$(dirname "$(dirname "$(dirname "$0")")")
# Or use the known skill directory structure
KIT_PATH="${HOME}/.claude/skills/work-summary-kit"Step 8: Create Output Directories
mkdir -p {output_dir}/daily
mkdir -p {output_dir}/weeklyStep 9: Validate
Run quick checks:
- Config readable: Read back the config and confirm it parses
- Git repos accessible: For each repo, verify the path exists and is a git repo
- Env vars set: Check all required env vars are non-empty
- Output dir writable: Touch a test file and remove it
Report results:
Setup complete! Your config is at ~/.config/work-summary/config.json
Enabled sources: Git, JIRA, Calendar Output directory: ~/work-summaries/ Template: daily-default
To generate your first daily summary, run: /daily-summary To set up automated scheduling, see: {kit_path}/scheduling/README.md
Warnings:
- SLACK_TOKEN not set (Slack will be skipped until configured)
Step 10: Offer Test Run
Ask:
Want me to run a quick test summary now? I'll generate today's summary with your new config.
If yes, invoke the daily-summary skill.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

