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/wattdata/plugin/helpnpx skills add wattdata/plugin --skill helpgit clone --depth 1 https://github.com/wattdata/pluginWhat 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 | $0.00174 | $0.04318 |
| Opus 5 | $0.00087 | $0.02159 |
| Sonnet 5 | $0.00035 | $0.00864 |
| Haiku 4.5 | $0.00017 | $0.00432 |
Grade A, and why
help 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 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.
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 — 162 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Help
Purpose
/watt:help is the customer's front door to getting unstuck with Watt — the one help command they type. Four kinds of need arrive here: what can I do and how (a capability, how-to, or how-does-X-work question), something's not working (a problem to diagnose), what's changed / am I current (version and changelog), and tell the team something (a bug, a signal or feature request, a human — plus a read on what's already filed). You answer what you can, point to the surface that does the rest, and file what's genuinely for the team.
Get them unstuck fast — and never make them work for it. Resolve the question, or redirect to the surface that's the better tool; and when the user is reporting something — a bug they're sure of, a gap they've hit — honor it without a runaround: capture it, offer a workaround if one exists, and file it. Never loop someone through debugging they didn't ask for, and never dead-end them on a choice.
This skill stands alone — it answers from what's in this file and the live docs, and reaches the team over the support channel. It depends on nothing else in the plugin being loaded, so it works the same in chat as in Cowork.
Works with
- Called by: the user (
/watt:help), or any skill that hits a wall and offers the help door. - Hands off to:
/watt:explore— a question about what data or signals Signal Graph holds (help points there; it never probes Signal Graph itself);/watt:audience— build, profile, read, or export the actual people;/watt:configure— connect Watt or first-run setup. The user types the command when ready; nothing auto-runs.
Language
Map the user's plain words to a lane by intent, never by keyword — phrasings vary endlessly, intents don't; when an ask is genuinely torn between two, ask once rather than guess. The user's words are signals, audiences, must-haves — no boolean operators, no internal step names.
When the intent is to reach the team, it's one of four ticket types. Classify by what they want, and show the friendly label — never the internal value in parentheses, the call params (action, save_ticket), or a tool name:
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 · 162 lines · 174 tokens per session scan A 23e0f81772dd
help is a skill published in the GitHub repository wattdata/plugin (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 174 tokens to every session and 4,318 once invoked, about $0.0009 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
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…