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 instructions/brujack/dotfiles/claude-mdgit clone --depth 1 https://github.com/brujack/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 | $0.28383 | $0.28383 |
| Opus 5 | $0.14191 | $0.14191 |
| Sonnet 5 | $0.05677 | $0.05677 |
| Haiku 4.5 | $0.02838 | $0.02838 |
Grade B, and why
dotfiles CLAUDE.md scanned grade B with 2 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- Ubuntu: `install_bats_linux()` in `lib/linux_shared.sh` — `sudo apt-get install -y bats` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
| `_RHN_CURL_BIN` | `_rhn_notify` | `curl` is resolved by name, but a test must capture the payload rather than send it. Every ntfy assertion in the suite reads this capture | How it starts
The opening of the file, as written. The whole thing — 1,071 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — dotfiles
Repository Overview
Personal development environment bootstrapping system for macOS and Linux (Ubuntu). Manages shell configs, tool installation, and symlink setup across multiple machine types.
Layout
dotfiles/
├── setup_env.sh # Main entry — sources lib/, dispatches workflows
├── Brewfile # Homebrew bundle (100+ formulae/casks; [HAS_*] tags are capability-gated)
├── lib/ # Shell libraries: constants, helpers, detect_env, macos, linux_shared,
│ # linux_ubuntu, developer, update_summary, workflows, git_hooks,
│ # launch_agents
├── config/ # profiles.sh (hostname→profile map); local.sh (machine overrides, git-ignored);
│ # hook_repos.sh (expected-repos list for the git-hooks sweep)
├── scripts/ # bootstrap_mac.sh, whats-new-*.sh, run-bash-coverage.sh,
│ # cadence-notify.sh
├── LaunchAgents/ # cadence.plist.template — one template, both weekly agents
├── powershell/ # Windows bootstrap: setup_windows.ps1, Pester tests, Makefile
├── tests/ # BATS tests: setup_env/, zshrc.d/, mocks/, helpers/
├── docs/ # ADRs, knowledge/, superpowers/, claude-code-new-features/,
│ # anthropic-new-features/
├── .github/workflows/ # CI: test, lint-macos, bash-coverage, powershell, secret-scan, auto-merge
├── .zshrc, .vimrc, .tmux.conf, .gitconfig_*, starship.toml, bruce.omp.json, profile.ps1
└── ubuntu_*_packages.txt, .ssh/
10-80-10 Execution Cycle
Sessions in this repo follow the 10-80-10 execution cycle defined in ai-config ADR-0009 (with the ADR-0010 wave-dispatch extension):
- Phase 1 (10%) — Architect.
brainstorming→writing-plans(emit per-task YAMLyaml-taskblocks withrole/model/tdd/acceptance/max_retries/files_touched/depends_on/parallel_group). Opus role. - Phase 2 (80%) — Execute.
subagent-driven-developmentruns iterate-until-green per task; FORBIDDEN list prevents gate cheating; wave-dispatch whenparallel_groupis declared. Sonnet/Haiku per task per the plan. - Phase 3 (10%) — Review.
finishing-a-development-branchchainspr-review→security-review→bug-scan→docs→learnings→ finish. Opus role.
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 · 1,071 lines · 28,383 tokens per session scan B 21f467a4da92
dotfiles CLAUDE.md is an instructions file published in the GitHub repository brujack/dotfiles (5 stars, last pushed 2d ago), licensed Apache-2.0. It adds 28,383 tokens to every session, about $0.1419 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
next.js AGENTS.md
Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.