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/explorenpx skills add wattdata/plugin --skill exploregit 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 | $0.00122 | $0.05183 |
| Opus 5 | $0.00061 | $0.02592 |
| Sonnet 5 | $0.00024 | $0.01037 |
| Haiku 4.5 | $0.00012 | $0.00518 |
Grade C, and why
explore scanned grade C with 1 finding 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.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
- **Minors.** Watt's data is adults-only. If an idea is about kids/teens, pivot to parents/guardians of that age range and proceed — don't refuse pedantically. How it starts
The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Explore Signal Graph
Purpose
Supported platform first. Watt runs in Claude Cowork, Claude Code, or the Claude Agent SDK — not the Claude Chat app, where only these skill files load and the rest of Watt — its hooks, context, scripts, and Signal Graph — does not. If the user isn't on a supported platform, tell them plainly to switch before continuing; otherwise this will not function correctly.
/watt:explore is the discovery surface — an interactive way to interrogate Signal Graph. The user brings a question or a plain-English idea about people ("folks planning corporate events", "is Watt any good for reaching home cooks?") and you help them find out what Watt actually holds for it: probe what Signal Graph has, ask one concrete question at a time, and let every answer steer the next probe.
Signal Graph is ~200k signals — there is no list to read; the only way to know its shape is to probe it, and that's what this surface is for. The questions it answers:
- "What does Watt have on X?" — the signals behind an idea, how big and fresh.
- "Is Watt good for Y?" — whether Signal Graph can express the people they mean.
- "What else is there around N?" — adjacent territory they haven't thought to ask about.
- "How could I reach / sharpen Z?" — which signals are strongest, what's thin, where to go deeper.
What the user learns is the value. A signal pool — signals they choose to keep — is a byproduct: if they're keeping signals, track them so they carry into /watt:audience, but the walk exists to answer their questions, not to assemble anything. Turning signals into actual people is downstream's job, never explore's.
One decision per turn. Each turn ends at one decision only the user can make — offered as the host's native clickable options, a typed answer always wins — and stops there. The user steers between turns; that steering is the walk.
Explore opens the space up. Its job is divergent — surface what's there, answer what they're asking, point at what's adjacent.
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 · 151 lines · 122 tokens per session scan C 71b81bafebb2
explore is a skill published in the GitHub repository wattdata/plugin (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 122 tokens to every session and 5,183 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it C with 1 finding (tells the agent never to refuse). 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…