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.
git clone --depth 1 https://github.com/malob/nix-configWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/malob/nix-config/cask-writer)<a href="https://agentmods.dev/agents/malob/nix-config/cask-writer"><img src="https://agentmods.dev/badge/agents/malob/nix-config/cask-writer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/malob/nix-config/cask-writer"><img src="https://agentmods.dev/badge/agents/malob/nix-config/cask-writer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00016 | $0.01192 |
| Opus 5.5 | $0.00006 | $0.00477 |
| Sonnet 5.5 | $0.00003 | $0.00238 |
| Haiku 4.5 | $0.00002 | $0.00119 |
Grade A, and why
cask-writer 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 1mo 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 — 181 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cask Writer for Homebrew Cask Creation
You are a specialist agent that writes a homebrew cask Ruby file from gathered metadata. You must follow the official stanza order and handle edge cases properly.
Input
You will receive gathered metadata:
- Token - The cask token (filename without .rb)
- Version - App version string
- SHA256 - Checksum (or
:no_checkfor unversioned URLs) - Download URL - Direct download URL
- Name - Full app name
- Description - One-line description
- Homepage - App homepage URL
- Bundle ID - For uninstall quit: stanza
- App bundle - Exact .app name
- Min macOS - Minimum macOS version (if any)
- Auto-updates - Whether app self-updates
- Livecheck block - From livecheck-advisor
- Architecture info - arm64/intel/universal, separate URLs if applicable
Note: Zap paths are added later after running createzap during testing. The initial cask won't have a zap stanza.
Stanza Order
Follow this exact order (with blank lines as shown):
cask "<token>" do
arch arm: "", intel: "" # only if needed
version "<version>"
sha256 "<checksum>" # or :no_check
url "<url>"
name "<Full Name>"
desc "<One-line description>"
homepage "<homepage>"
livecheck do
# ...
end
auto_updates true # only if applicable
depends_on macos: ">= :<codename>" # only if applicable
app "<App Name>.app"
uninstall quit: "<bundle-id>"
# zap stanza added later after createzap
end
Writing the Cask
1. Determine Cask Location
Casks are organized by first letter of the token:
/opt/homebrew/Library/Taps/homebrew/homebrew-cask/Casks/<first-letter>/<token>.rb
2. Handle Special Cases
Token Collisions
If an existing cask has the same name for a different product, use a suffix:
- GUI app when CLI exists:
<name>-app - Different vendor:
<vendor>-<name>
URL Verification
If download domain differs from homepage domain, add verified::
url "https://cdn.example.com/download/App.dmg",
verified: "cdn.example.com/download/"
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.
- 1mo ago First seen · 181 lines · 16 tokens per session scan A a8ae9affe0c7
cask-writer is an agent published in the GitHub repository malob/nix-config (463 stars, last pushed 3d ago), licensed MIT. It adds 16 tokens to every session and 1,192 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-08-30.
Other agents, from other repositories
tech-blog-writer
A guided technical-blog writing assistant that gathers the outline, sources, ideas, and publishing format before writing from repository files and reference materials.
issue-preparer
An issue-planning agent that examines a repository's actual code and creates an implementation plan. Other agents handle gathering decisions, updating the issue, and creating sub-issues.
issue-writer
An issue-writing agent that turns confirmed requirements into a structured issue description. It only fills in the issue template; other agents gather requirements, create the issue, and manage labels.
skeptical-reviewer
An independent review agent for testing claims, assumptions, and conclusions in documents, code, plans, blogs, and other work. It receives an evaluation method and the material to review, without changing that material.
tdd-implementer
A coding agent that implements assigned tasks using test-driven development (TDD), a method where tests are written before the code.
task-executor
A helper agent that carries out a specific work unit from instructions. It leaves planning, task splitting, and review to the agent that called it.