hmem-new-project

name	hmem-new-project
description	Create a new P-entry (project) in hmem via create_project — handles schema, section setup, and O-entry linking. Use when the user asks to add/register/track a new project ('neues Projekt', 'P-Entry erstellen'), or before any write_memory with prefix='P'.

/hmem-new-project — New Project Entry

Uses the create_project tool to set up a complete project with one call.

Step 1: Does a codebase exist?

Ask:

"Gibt es bereits eine Codebase? Wenn ja, in welchem Verzeichnis?"

If yes: Scan the directory (README, package.json, CLAUDE.md, etc.) to extract name, tech stack, description, entry points, key modules. Use findings to fill the create_project parameters and later append Codebase details.

If no: All info comes from the user's answers.

Step 2: Quick questions (one at a time)

Ask these in order. Skip what the codebase scan already answered.

1. Name — "Wie soll das Projekt heißen?"
2. Stack — "Welche Technologien?"
3. One-liner — "Beschreib das Projekt in einem Satz"
4. Goal — "Was ist das Hauptziel?"
5. Who — "Wer nutzt es?"
6. Repo — "Repo-Pfad oder URL?"
7. Deployment — "Wie wird es deployed?"

Stop early if the user says "reicht" or "das war's".

Step 3: Create the project

create_project({
  name: "...",
  tech: "...",
  description: "...",
  goal: "...",
  repo: "...",
  audience: "...",
  deployment: "...",
  tags: ["#lang", "#framework"]
})

This creates:

• P00XX with sections from the configured schema (or 9 R0009 defaults if no schema)
• O00XX matching O-entry for session logging (if createLinkedO: true in schema, or always when no schema)

Step 4: Fill in details

If a codebase was scanned, append the findings. The Codebase node (.2) follows a strict schema:

Codebase node structure

• L3 — Pipeline (first child): Data flow overview — entry → moduleA → moduleB → storage
• L3 — Modules: One node per source file. Title = filename, body = purpose + src/file.ts
• L4 — Functions: Under each module, one node per exported function/class. Title = full TypeScript signature, body = one-line description + src/file.ts

append_memory(id="P00XX.2", title="Pipeline", body="src/cli.ts → moduleA.ts → DB")
append_memory(id="P00XX.2", title="moduleA.ts", body="Core logic handler. src/moduleA.ts")
append_memory(id="P00XX.2.N", title="doThing(x: string): Promise<boolean>", body="Does the thing. src/moduleA.ts")

L4 is critical. Without function signatures, every agent must read source files for every task. Fill L4 for all exported functions when creating a new project — dispatch a subagent to extract them if the codebase is large.

• L5 — Extended Notes (optional, use selectively): Add under an L4 node only when it adds real value:

  • Usage example: write("P", { content: "Title\n\nBody", tags: ["#foo"] })
  • Important caveats: "triggers sync push", "returns null if not found — never throws"
  • Parameter details: field list for complex option types like ReadOptions

  Not every function needs L5 — only where the signature alone leaves agents guessing.

append_memory(id="P00XX.3", content="Installation: npm install\n\n...")

Step 5: Link related entries

Search for existing entries that relate to this project and add links:

read_memory(search="project keywords")
update_memory(id="P00XX", content="...", links=["T0044", "L0095"])

Then show the result: load_project(id="P00XX")