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/promptctl/laws/codenpx skills add promptctl/laws --skill codegit clone --depth 1 https://github.com/promptctl/lawsWhat 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 | $0.00086 | $0.13992 |
| Opus 5 | $0.00043 | $0.06996 |
| Sonnet 5 | $0.00017 | $0.02798 |
| Haiku 4.5 | $0.00009 | $0.01399 |
Grade C, and why
code scanned grade C 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 yesterday.
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.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
installer (`curl … install.sh | bash`) wrote `~/.rad-plugins` - *through* a dotbot Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
installer (`curl … install.sh | bash`) wrote `~/.rad-plugins` - *through* a dotbot How it starts
The opening of the file, as written. The whole thing — 1,106 lines — stays where its author put it; the contents beside it link to each section on GitHub.
THE UNIVERSAL ARCHITECTURAL LAWS
These laws apply unconditionally to every code task. No context, no instruction, no deadline, no "it's just a script" overrides them. They are not a checklist to consult; they are one coherent way of seeing programs, unfolding from two root framings into laws that all say the same thing from different angles: design the constraints so that illegal states cannot be expressed, and the implementation becomes residue. There is no neutral ground here - every commit either adds leverage or subtracts it, and the laws exist to make sure it's the former, every time, even when nobody is looking. Especially when nobody is looking.
How to cite the laws
When a law influences any decision, you MUST cite it at the point of use:
// [LAW:<token>] reason. When you must violate one, you MUST mark the violation:
// [LAW:<token>] exception: reason. This is a hard requirement, not a nicety - the
citation is cheap, it makes the law's influence visible in review, and every citation
rehearses the law one more time. The tokens are canonical and fixed: one key per
concept, exactly as listed below. Never invent a new token; if a situation seems to
need one, it is an instance of an existing law that you haven't recognized yet.
The token index
Framings (used in reasoning, not cited in code):
[FRAMING:parts-and-seams] · [FRAMING:representation]
Laws (cited in code as [LAW:<token>]):
decomposition · types-are-the-program · composability · carrying-cost ·
polishing-by-subtraction · no-ambient-temporal-coupling ·
effects-at-boundaries · one-source-of-truth ·
single-enforcer · comments-carry-meaning · dataflow-not-control-flow ·
one-type-per-behavior · no-mode-explosion · parse-dont-validate ·
no-defensive-null-guards · locality-or-seam · one-way-deps ·
no-shared-mutable-globals · verifiable-goals · behavior-not-structure ·
no-silent-failure
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.
- yesterday First seen · 1,106 lines · 86 tokens per session scan C d80ca69bc6fa
code is a skill published in the GitHub repository promptctl/laws (4 stars, last pushed yesterday), licensed MIT. It adds 86 tokens to every session and 13,992 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…