borklog

Scaffold per-project memory files and insert a CLAUDE.md section.

1|Updated May 20, 2025
One-click install
npx skills add https://github.com/vessux/dotfiles --skill borklog
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: borklog
Source: https://github.com/vessux/dotfiles/tree/main/umbel/skills/local/borklog
Command: npx skills add https://github.com/vessux/dotfiles --skill borklog

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Scaffolds per-project memory by creating backlog and worklog files and injecting a CLAUDE.md section with rules and schemas. Idempotent re-run for template updates.

Core Features & Use Cases

  • Creates docs/backlog.jsonl and docs/worklog.jsonl in the project root if missing.
  • Inserts a Project memory section into CLAUDE.md with versioned markers for future updates.
  • Designed for a solo, single-agent workflow to avoid multi-author conflicts.

Quick Start

Run the borklog skill in your project to scaffold the backlog/worklog files and CLAUDE.md section.

Frequently Asked Questions about borklog

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I set up per-project memory files for Claude Code?

To set up per-project memory, you can run a one-shot scaffolder that creates docs/backlog.jsonl and docs/worklog.jsonl, and injects a rules section into CLAUDE.md.

What is the best way to scaffold a backlog and worklog in my project root?

Scaffolding a backlog and worklog involves generating docs/backlog.jsonl and docs/worklog.jsonl files, alongside injecting versioned schema markers into CLAUDE.md for future template updates.

Can I re-run a CLAUDE.md scaffolder to update rules without overwriting my project docs?

Yes, idempotent scaffolding tools can re-run safely to update CLAUDE.md rules within versioned markers without overwriting existing backlog or worklog data files.

Does per-project memory scaffolding support multi-agent workflows?

Per-project memory scaffolding is designed specifically for solo, single-agent workflows to avoid multi-author conflicts when writing to the backlog and worklog files.

How do I initialize project memory using a jsonl format for task tracking?

You initialize project memory by triggering explicit commands like init or scaffold, which creates jsonl formatted files for tracking tasks and injects schemas into CLAUDE.md.

Why won't my project memory scaffolder run automatically when I start a session?

Project memory scaffolding will not run automatically because it requires an explicit initialization request, such as a direct command or slash invocation, to create the necessary files.