The agent will scan the project and automatically generate a CLAUDE.md file, along with the necessary skills and AI agents.
Paste this entire document into Claude Code at the start of any new project. Work through every phase in order. Do not skip phases. Use WebFetch, Bash, Read, Write, and Edit tools freely throughout.
| # .bunfig.toml | |
| [install] | |
| minimumReleaseAge = 604800 # 7 days in seconds |
| # syntax=docker/dockerfile:1 | |
| # Download, build, and run: | |
| # curl -fsSL https://gist.githubusercontent.com/nikarh/6381b3d43eb92b2479f06482fbf8e02f/raw/Dockerfile -o /tmp/Dockerfile && docker build --build-arg UID="$(id -u)" --build-arg GID="$(id -g)" --build-arg USERNAME="${USER:-dev}" -t agents:latest -f /tmp/Dockerfile /tmp && docker run --name agent -it -v "$(pwd):/workspace" -e GIT_AUTHOR_NAME="$(git config --get user.name)" -e GIT_AUTHOR_EMAIL="$(git config --get user.email)" -e GIT_COMMITTER_NAME="$(git config --get user.name)" -e GIT_COMMITTER_EMAIL="$(git config --get user.email)" agents:latest | |
| FROM archlinux:latest | |
| SHELL ["/bin/bash", "-o", "pipefail", "-c"] | |
| ARG UID=1000 | |
| ARG GID=1000 |
| name | generate-agents | ||
|---|---|---|---|
| description | Analyzes any project and scaffolds a complete .agents/ folder with AGENTS.md, sync script, and starter skills tailored to the detected tech stack. Use when bootstrapping AI agent configuration for a new or existing repository. | ||
| metadata |
|
You are a meta-scaffolding expert. Analyze the project's codebase and produce a complete .agents/ folder that gives AI coding agents project-specific context, conventions, and reusable skills.
| #!/usr/bin/env zsh | |
| # Multi-project worktree manager with Claude support | |
| # | |
| # ASSUMPTIONS & SETUP: | |
| # - Your git projects live in: ~/projects/ | |
| # - Worktrees will be created in: ~/projects/worktrees/<project>/<branch> | |
| # - New branches will be named: <your-username>/<feature-name> | |
| # | |
| # DIRECTORY STRUCTURE EXAMPLE: | |
| # ~/projects/ |
| name | api-developer | ||
|---|---|---|---|
| description | Implement API endpoints following team conventions | ||
| skills |
|
Implement API endpoints. Follow the conventions and patterns from the preloaded skills.
You can scale a SQLite database to multiple GByte in size and many concurrent readers by applying the below optimizations.
(some are applied permanently, but others are reset on new connection)
pragma journal_mode = WAL;
Instead of writing directly to the db file, write to a write-ahead-log instead and regularily commit the changes. Allows multiple concurrent readers, and can significantly improve performance.
sudo dnf install perfetto (official RPM, currently perfetto-49.0-1.fc42 and perfetto-49.0-5.fc45 in Rawhide), which installs traced, traced_probes, the perfetto CLI, and ready-made systemd units.tracebox from get.perfetto.dev (a single statically linked binary) or build from source.trace_processor wrapper from get.perfetto.dev or python3 -m pip install perfetto; the AL2023 glibc (exactly glibc-2.34-52.amzn2023) meets the prebuilt trace_processor_shell requirement (GLIBC_2.34) with zero headroom.ui.perfetto.dev, which processes the trace entirely in your browser (nothing is uploaded unless you click "Share", which only works for Googlers); you can self-host the same UI on any static web server.