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 skills add TanStack/devtools --skill devtools-bidirectionalgit clone --depth 1 https://github.com/TanStack/devtoolsWrote 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/skills/tanstack/devtools/devtools-bidirectional)<a href="https://agentmods.dev/skills/tanstack/devtools/devtools-bidirectional"><img src="https://agentmods.dev/badge/skills/tanstack/devtools/devtools-bidirectional/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/skills/tanstack/devtools/devtools-bidirectional"><img src="https://agentmods.dev/badge/skills/tanstack/devtools/devtools-bidirectional.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00059 | $0.03297 |
| Opus 5.5 | $0.00024 | $0.01319 |
| Sonnet 5.5 | $0.00012 | $0.00659 |
| Haiku 4.5 | $0.00006 | $0.00330 |
Grade A, and why
devtools-bidirectional 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 — 473 lines — stays where its author put it; the contents beside it link to each section on GitHub.
devtools-bidirectional
Prerequisite: Read and understand the
devtools-event-clientskill first. This skill builds onEventClient, its event map types,emit()/on()API, pluginId namespacing, connection lifecycle, and singleton pattern. Everything here assumes you already have a workingEventClientinstance.
Two-way communication between your application and a TanStack Devtools panel using EventClient. The same client instance handles both directions: the app emits observation events that the panel listens to, and the panel emits command events that the app listens to.
Core Concept
EventClient is not unidirectional. Both emit() and on() work from either side -- application code or panel code -- on the same shared event bus. The direction is a convention you establish through your event map design, not a limitation of the API.
App code calls: client.emit('state-update', ...) // observation
Panel code calls: client.on('state-update', ...) // observation
Panel code calls: client.emit('set-state', ...) // command
App code calls: client.on('set-state', ...) // command
Core Patterns
1. App-to-Devtools Observation
The app emits state changes. The panel listens and renders.
Event map and client (shared module):
import { EventClient } from '@tanstack/devtools-event-client'
type CounterEvents = {
// Observation: app -> panel
'state-update': { count: number; updatedAt: number }
}
class CounterDevtoolsClient extends EventClient<CounterEvents> {
constructor() {
super({
pluginId: 'counter-inspector',
enabled: process.env.NODE_ENV !== 'production',
})
}
}
export const counterClient = new CounterDevtoolsClient()
App side -- emit on state changes:
import { counterClient } from './counter-devtools-client'
function increment() {
count += 1
counterClient.emit('state-update', {
count,
updatedAt: Date.now(),
})
}
Panel side -- listen and display:
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 · 473 lines · 59 tokens per session scan A e2a7db665f88
devtools-bidirectional is a skill published in the GitHub repository TanStack/devtools (506 stars, last pushed today), licensed MIT. It adds 59 tokens to every session and 3,297 once invoked, about $0.0002 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 skills, from other repositories
vite-browser-core-debug
Core troubleshooting workflow for Vue/React/Svelte Vite apps using vite-browser. Use this for first-pass debugging when the user reports broken UI behavior, wrong component state, router/store confusion, or a vague "page is broken" symptom and the dominant failure mode is not yet known.
migrate-oxlint
Guide for migrating a project from ESLint to Oxlint. Use when asked to migrate, convert, or switch a JavaScript/TypeScript project's linter from ESLint to Oxlint.
react-doctor
Use when finishing a feature, fixing a bug, before committing React code, or when the user types /doctor, asks to scan, triage, or clean up React diagnostics. Covers lint, accessibility, bundle size, architecture. Includes a regression check and a full local-triage workflow that fetches the canonical playbook.
d3k
Use when the user asks to use d3k, run/dev/test/debug a web project with d3k, or reproduce a browser issue. Own the runtime: reuse or background-start d3k non-interactively, wait for readiness, use its project-stable managed Chrome profile, and inspect unified browser/server evidence.
dev-fix
Unified developer workflow for fixing bugs. Analyzes issue-tracker context, cross-checks docs/code, proposes a solution, implements the fix, verifies locally, and delivers a PR/MR.
incident-hotfix
Mitigate a production incident or urgent regression first, then route to root-cause remediation and a postmortem.