agentleFS
jalmulla2jalmulla2/claude-plugin-second-brain-kit/skills/new-project/SKILL.md

new-project

Skill0 starsChanged 41 days agoMITView on GitHub

Interviews the user, then creates a project folder and PROJ overview hub and registers it so later sessions know it exists. Use when the user says new project or wants to start something new.

51 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use checkpoint, prd and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • genomewalkerSkill

    checkpoint

    Save a checkpoint of task status, decisions, active files, and next steps

    Also in 3 projects that use this

  • GitHubSkill

    prd

    Generate high-quality Product Requirements Documents (PRDs) for software systems and AI-powered features. Includes executive summaries, user stories, technical specifications, and risk analysis.

    Also in 3 projects that use this

  • sunnyclinches44Skill

    explain

    Explains code using real-world analogies and ASCII diagrams

    Also in 3 projects that use this

  • SentrySkill

    spec

    Produce a verified spec — problem, desired outcome, and acceptance criteria — before building. Use when given a GitHub/Linear issue link, when starting from a fuzzy in-conversation idea, or asked to "figure out what to do" / "scope this" / "what should I build" — whenever the *what* isn't yet pinned down. Grills the ambiguities, checks the premise against the code, and produces acceptance criteria you sign off on — then hands to design-first.

    Also in 3 projects that use this

  • GoogleSkill

    domain-modeling

    Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, record an architectural decision, or when another skill needs to maintain the domain model.

    Also in 4 projects that use this

  • SentrySkill

    review

    Fixture review skill.

    Also in 5 projects that use this

  • +1 more from getsentry

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 2 are distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 48 edited versions we read keep these.

  • 2

    Distinctive

    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
  1. Very specific:

    If personal.md exists in this folder, read it first.

    line 9
  2. Distinctive:

    Not for an ongoing responsibility with no end state.

    line 17
  3. Very specific:

    Follow the language rule in CLAUDE.md (its ## Language block, if it has one).

    line 21
  4. Very specific:

    No project key on the hub: Every *other* file in the folder points at it with project: "[[PROJ <Project Name> Overview]]".

    line 101
  5. Very specific:

    MEMORY.md says it is generated from ## State blocks: add a State block to the hub directly after the frontmatter, then run the …

    line 110
  6. Distinctive:

    If any check fails, return to the step that produced it, fix it, and check again.

    line 129

“Common” is measured: kept by at least 6 in 10 of the 48 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

New Project

If personal.md exists in this folder, read it first; it overrides the defaults above.

Interviews the user about a new project, creates its folder and hub file, and registers it so good-morning and future sessions know it exists. Nothing is written until the interview is done.

When to use

  • The user says "new project", "start a project", "create a project", "add a project" or "I want to work on something new".
  • good-morning hands over because the user wants to start something new.
  • Not for an ongoing responsibility with no end state. That is an area, not a project (Step 0).

Language

Follow the language rule in CLAUDE.md (its ## Language block, if it has one). Write the prose (goal, why, outcomes, problems) in the conversation's current language. Frontmatter keys, section headings, the folder name and the PROJ <Name> Overview.md filename pattern stay English regardless.

Checklist

- [ ] 0. Is it a project? Does it already exist?
- [ ] 1. Interview, one question at a time
- [ ] 2. Create the folder and the hub
- [ ] 3. Register the project
- [ ] 4. Verify (Step 4); fix and re-check if anything fails
- [ ] 5. Confirm and offer to start

Step 0 — Is it a project, and is it new?

Projects end; areas don't. A project is time-bounded and has an end state, something that will be finished. If there is no answer to "what does done look like?", it is an ongoing area of responsibility. Say so instead of creating a folder that can never close. Ask this of yourself after the "done" question, not out loud.

Already started? The user may have started it in another conversation. Search the vault for the keyword before creating anything (find . -iname '*<keyword>*' from the vault root, or the area's index note if the vault has one).

Name collision. No PROJ <Project Name> Overview.md may already exist anywhere in the vault. Two notes with the same basename make every [[link]] to them ambiguous. If it collides, tell the user and ask for a different name.

Step 1 — Interview

One question per message. Send it, stop, wait for the answer. Don't batch the questions or list them up front. If the user answers several at once, confirm what you captured and resume from the first unanswered one.

  1. Name — what's it called?
  2. Goal — what is it trying to accomplish? One sentence is fine.
  3. Why — why does it matter? The real reason.
  4. Done — what will exist when it has succeeded?
  5. Open problems — anything they already know they'll have to solve? Fine if not.
  6. Where it belongs — only if the vault has more than one project root. With a 03 Life/ folder: work or personal? With area folders: which area? Infer it from the answers and confirm rather than asking cold.

Step 2 — Create the folder and the hub

Where it goes:

  • work, or the vault has no 03 Life/ → 02 Projects/<Project Name>/
  • personal → 03 Life/Projects/<Project Name>/ (create 03 Life/Projects/ if it isn't there yet)

The folder starts with one file, the project's hub: PROJ <Project Name> Overview.md. It is what Claude reads first whenever the project comes up, and it ties the project's files together. Never start the filename with [; Obsidian can't link to it.

Template:

---
title: <Project Name> Overview
description: <one sentence: what this project is and what finishing it looks like>
author: claude
type: overview
date: YYYY-MM-DD
update date: YYYY-MM-DD
status: active
tags: []
---

## Goal
<their goal answer>

## Why
<their why answer>

## Tangible Outcomes
- <outcome 1>
- <outcome 2>

## Open Problems
1. <problem>
<or, if they had none: 1. (to be defined — we'll work these out as we go)>

## Key Files
| File | Purpose |
|------|---------|

_No files yet — a row is added here every time a file is created in this project folder._

## Links Out
_Nothing yet._
  • description is what an agent (and any index) reads to judge relevance without opening the note. Make it say what the project is, not the title restated.
  • No project key on the hub. Every other file in the folder points at it with project: "[[PROJ <Project Name> Overview]]"; that is how the hub collects them as backlinks. A file can't belong to itself.
  • ## Key Files starts empty and doesn't stay empty. Whenever a file is created in the project folder, by any skill or by ordinary work, set its project key and add a row in the same action: | [[Filename Without Extension]] | <one phrase: why this file exists> |. Delete the placeholder line with the first row. end-of-day adds any row that was missed, as a safety net.
  • ## Links Out is a labelled list of deliberate relationships to notes outside the folder (- Depends on: [[PROJ Other Project Overview]], - Background: [[Some Note]], external URLs). Filename-only wikilinks. Add one only when a real relationship exists.

Step 3 — Register the project

This step is what makes future sessions aware of the project. Follow whichever applies:

  • CLAUDE.md has an ## Active Projects table: add a row, | <Project Name> | [[PROJ <Project Name> Overview]] |. The link is filename-only, so the row survives the project moving between roots. Don't change the table's structure, and don't add a folder map anywhere in CLAUDE.md; a stale map is worse than none.
  • MEMORY.md says it is generated from ## State blocks: add a State block to the hub directly after the frontmatter, then run the generator its header names. Don't edit MEMORY.md or CLAUDE.md by hand.

    ## State
    - **Status:** active
    - **Where it stands:** <one sentence: what this is for, where it is>
    - **Blocked on:** nothing
    - **Next:** <the single next concrete action>

    State, never events: four lines, no past-tense history.

  • The vault keeps generated index notes: regenerate the owning one with the tool that builds it. Never hand-edit an index.

Step 4 — Verify before confirming

  • The hub exists at the path from Step 2, and its basename is unique in the vault.
  • Its frontmatter parses (the --- fences are intact) and it has no project key.
  • The project is registered by every route in Step 3 that applies: the table row, or the State block plus a rebuilt MEMORY.md with the project's line in it.

If any check fails, return to the step that produced it, fix it, and check again.

Step 5 — Confirm and offer to dive in

Tell the user, briefly: the folder and hub are created and where, and the project is registered so later sessions will know it. Then ask: "Want to dive into one of the open problems now, or save it for later?"

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.