start

Q: How do I install start?

Run npx skills add https://github.com/anthropics/knowledge-work-plugins --skill start in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/start/ (personal) or .claude/skills/start/ (project).

by anthropics OfficialProductivity & Automation

Initialize the productivity system and open the dashboard. Use when setting up the plugin for the first time, bootstrapping working memory from your existing task list, or decoding the shorthand (nicknames, acronyms, project codenames) you use in your todos.

Installs: 4.9K
Last 24h: +9
Trending rank: #8975
Security: Pass

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/anthropics/knowledge-work-plugins --skill start

01
Install
Run the command in your project. The CLI asks which agents to install for.
02
Describe the task
Ask your agent for something this skill covers. It loads start automatically when relevant.
03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.

Security audits

Independent scans of the skill's instructions and bundled files.

Gen Agent Trust Hub PassSAFE ·This skill initializes a productivity dashboard by analyzing user tasks, emails, and messages to build a 'memory' system. It includes security considerations regarding the ingestion of untrusted data from external sources, which is a standard surface for indirect prompt injection. These elements are consistent with the skill's initialization and bootstrapping purpose.

Socket PassNo alerts

Snyk PassLOW ·Risk: LOW · No issues

Runlayer PassLOW ·1/1 file flagged

ZeroLeaks PassNONE ·Score: 93/100 · 2 sections analyzed

SKILL.md

The instructions your agent reads when the skill is triggered.

Start Command

If you see unfamiliar placeholders or need to check which tools are connected, see CONNECTORS.md.

Initialize the task and memory systems, then open the unified dashboard.

Instructions

1. Check What Exists

Check the working directory for:

TASKS.md — task list
CLAUDE.md — working memory
memory/ — deep memory directory
dashboard.html — the visual UI

2. Create What's Missing

If TASKS.md doesn't exist: Create it with the standard template (see task-management skill). Place it in the current working directory.

If dashboard.html doesn't exist: Copy it from ${CLAUDE_PLUGIN_ROOT}/skills/dashboard.html to the current working directory.

If CLAUDE.md and memory/ don't exist: This is a fresh setup — after opening the dashboard, begin the memory bootstrap workflow (see below). Place these in the current working directory.

3. Open the Dashboard

Do NOT use open or xdg-open — in Cowork, the agent runs in a VM and shell open commands won't reach the user's browser. Instead, tell the user: "Dashboard is ready at dashboard.html. Open it from your file browser to get started."

4. Orient the User

If everything was already initialized:

Dashboard open. Your tasks and memory are both loaded.
- /productivity:update to sync tasks and check memory
- /productivity:update --comprehensive for a deep scan of all activity

If memory hasn't been bootstrapped yet, continue to step 5.

5. Bootstrap Memory (First Run Only)

Only do this if CLAUDE.md and memory/ don't exist yet.

The best source of workplace language is the user's actual task list. Real tasks = real shorthand.

Ask the user:

Where do you keep your todos or task list? This could be:
- A local file (e.g., TASKS.md, todo.txt)
- An app (e.g. Asana, Linear, Jira, Notion, Todoist)
- A notes file

I'll use your tasks to learn your workplace shorthand.

Once you have access to the task list:

For each task item, analyze it for potential shorthand:

Names that might be nicknames
Acronyms or abbreviations
Project references or codenames
Internal terms or jargon

For each item, decode it interactively:

Task: "Send PSR to Todd re: Phoenix blockers"

I see some terms I want to make sure I understand:

1. **PSR** - What does this stand for?
2. **Todd** - Who is Todd? (full name, role)
3. **Phoenix** - Is this a project codename? What's it about?

Continue through each task, asking only about terms you haven't already decoded.

6. Optional Comprehensive Scan

After task list decoding, offer:

Do you want me to do a comprehensive scan of your messages, emails, and documents?
This takes longer but builds much richer context about the people, projects, and terms in your work.

Or we can stick with what we have and add context later.

If they choose comprehensive scan:

Gather data from available MCP sources:

Chat: Recent messages, channels, DMs
Email: Sent messages, recipients
Documents: Recent docs, collaborators
Calendar: Meetings, attendees

Build a braindump of people, projects, and terms found. Present findings grouped by confidence:

Ready to add (high confidence) — offer to add directly
Needs clarification — ask the user
Low frequency / unclear — note for later

7. Write Memory Files

From everything gathered, create:

CLAUDE.md (working memory, ~50-80 lines):

# Memory

## Me
[Name], [Role] on [Team].

## People
| Who | Role |
|-----|------|
| **[Nickname]** | [Full Name], [role] |

## Terms
| Term | Meaning |
|------|---------|
| [acronym] | [expansion] |

## Projects
| Name | What |
|------|------|
| **[Codename]** | [description] |

## Preferences
- [preferences discovered]

memory/ directory:

memory/glossary.md — full decoder ring (acronyms, terms, nicknames, codenames)
memory/people/{name}.md — individual profiles
memory/projects/{name}.md — project details
memory/context/company.md — teams, tools, processes

8. Report Results

Productivity system ready:
- Tasks: TASKS.md (X items)
- Memory: X people, X terms, X projects
- Dashboard: open in browser

Use /productivity:update to keep things current (add --comprehensive for a deep scan).

Notes

If memory is already initialized, this just opens the dashboard
Nicknames are critical — always capture how people are actually referred to
If a source isn't available, skip it and note the gap
Memory grows organically through natural conversation after bootstrap

FAQ

What does the start skill do?

How do I install start?

Run npx skills add https://github.com/anthropics/knowledge-work-plugins --skill start in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/start/ (personal) or .claude/skills/start/ (project).

Which AI agents can use start?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is start safe to install?

Security partners have reviewed it: Gen Agent Trust Hub passed; Socket passed; Snyk passed; Runlayer passed; ZeroLeaks passed. As with any skill, read the SKILL.md and bundled scripts before installing.

Who maintains start?

It is published by anthropics in anthropics/knowledge-work-plugins, and it is an official skill from the maker of the technology it covers. It has 4,909 recorded installs.