Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/urmzd/dotfiles/write-code-portfolionpx skills add urmzd/dotfiles --skill write-code-portfoliogit clone --depth 1 https://github.com/urmzd/dotfilesWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00148 | $0.00788 |
| Opus 5 | $0.00074 | $0.00394 |
| Sonnet 5 | $0.00030 | $0.00158 |
| Haiku 4.5 | $0.00015 | $0.00079 |
Grade A, and why
write-code-portfolio scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Portfolio Dev Setup
Personal, opinionated tooling that backs the patterns in write-code. This is not portable. Other users on Bitwarden + asdf + bash will not want any of this.
Composable Toolsets (Nix Flakes)
13 reusable per-language shells combined via flake.nix outputs. Each shell pins a single toolchain (rust-toolchain, go, node, python with uv, etc.). Projects compose the shells they need; nothing pollutes the global env.
- Each language shell exposes a
devShelland alib.toolsset for downstream composition - Avoid
nix-direnvhere -- vanilla direnv + ause flakeline keeps cold-start fast and predictable on macOS
Machine Polymorphism (chezmoi)
Dotfile templates branch on host attributes:
{{ if .is_macos -}}
# macOS-only block (Homebrew bundle, defaults write, ...)
{{ end -}}
{{ if and .is_personal (not .is_work) -}}
# personal-only block (1Password personal vault, side projects, ...)
{{ end -}}
Attributes come from .chezmoi.toml.tmpl (is_macos, is_linux, is_personal, is_work). Keep templates small and the booleans flat; deeply nested {{ if }} blocks are a smell.
Powerlevel10k
Zsh prompt with instant prompt enabled and cache invalidation tied to plugin dir mtime. Source of truth: dot_p10k.zsh in the chezmoi source. Avoid editing the rendered ~/.p10k.zsh directly -- it is overwritten on next chezmoi apply.
Neovim (HEAD)
Lua config under private_dot_config/nvim/. Highlights:
- LSP set:
ty,ts_ls,lua_ls,gopls,rust_analyzer,jsonls,yamlls,bashls,jdtls - Subprocess-safe: detects non-interactive sessions and disables clipboard provider (so
nvim -c '...' +qin scripts doesn't hang on macOS pbcopy) - Plugin manager: lazy.nvim, pinned via
lazy-lock.json - 24-hour
compinitcache; auto fpath discovery picks up new completions on next shell start
Why split this out
The portable write-code skill answers "what should this codebase look like?" That question is the same for everyone in the org. This skill answers "how is my personal machine wired up?" -- a question only I should be making decisions for, so the router shouldn't pick it for general code-writing requests.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 62 lines · 148 tokens per session scan A d6f451f1e0e0
write-code-portfolio is a skill published in the GitHub repository urmzd/dotfiles (3 stars, last pushed 20d ago), licensed Apache-2.0. It adds 148 tokens to every session and 788 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
benchling-integration
Benchling Python SDK and REST API integration for registry entities, inventory, ELN entries, workflows, Benchling Apps, and Data Warehouse queries. Use when automating lab data with benchling-sdk or the v2 API.
astropy
Core Python library for astronomy and astrophysics workflows that need Astropy APIs, including units/quantities, coordinates, FITS I/O, tables, time systems, WCS, and cosmology. Use when implementing or debugging astronomical data analysis code with Astropy.
esm
Comprehensive toolkit for EvolutionaryScale protein language models including ESM3 (generative multimodal design across sequence, structure, and function) and ESM C (efficient embeddings). Use for protein sequence/structure/function tasks, inverse folding, embeddings, variant design, and ESMFold2 structure prediction…
create-pr
Rebase from the latest origin/main, squash the commits from it, and then create a PR on github with intelligent commit messages based on staged changes.
mcp-notion-usage-guide
MCP Notion工具使用指南，包含常见问题解决方案和最佳实践。使用当： (1) 访问Notion数据库view URL出现"URL type view not currently supported"错误， (2) 需要获取数据库schema和表结构信息， (3) 查询数据库中的条目内容， (4) 创建页面时MULTISELECT字段值不存在导致失败， (5) 需要更新数据库schema添加新选项， (6) 开发需要集成Notion数据的自动化工作流， (7) querydatasources/query-database-view返回Business Plan要求错误， (8) 需要在无Business…
scan-commit
Automatically scan the repository for unstaged/untracked changes, group them into logical commits using hunk-level analysis, stage and commit them following Conventional Commits.