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 labsai/EDDI --skill eddi-datagit clone --depth 1 https://github.com/labsai/EDDIWrote 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/labsai/eddi/eddi-data)<a href="https://agentmods.dev/skills/labsai/eddi/eddi-data"><img src="https://agentmods.dev/badge/skills/labsai/eddi/eddi-data/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/labsai/eddi/eddi-data"><img src="https://agentmods.dev/badge/skills/labsai/eddi/eddi-data.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 | $0.00046 | $0.00902 |
| Opus 5.5 | $0.00018 | $0.00361 |
| Sonnet 5.5 | $0.00009 | $0.00180 |
| Haiku 4.5 | $0.00005 | $0.00090 |
Grade A, and why
eddi-data 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 16d 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 — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.
EDDI Manager data & plumbing
AGENTS.md is the authority; this is the working summary for building a screen.
Routing
React Router v7, declarative mode (no data router). Routes are declared in src/app.tsx;
pages render through AppLayout's <Outlet />. Resource detail routes carry the version as
a query param: /manage/channels/:id?version=2. src/__tests__/route-integrity.test.ts
asserts every nav target resolves — add the route and the sidebar entry together.
Server state — TanStack Query v5
One hook file per domain in src/hooks/ (use-channels.ts, use-agents.ts, …). Pages
never call the API directly:
const { data: channels, isLoading, error, refetch } = useEnrichedChannelDescriptors();
const deleteMutation = useDeleteChannel();
await deleteMutation.mutateAsync({ id, version });
Filtering/search is client-side useMemo over the query result when the list is small —
that is the existing pattern, not a server round-trip per keystroke.
API calls
Use ApiClient (src/lib/api-client.ts) for ordinary JSON — it injects the Keycloak auth
header. Base URL is always window.location.origin; never hardcode. Raw fetch is only
justified for SSE, binary/blob and text/plain bodies, and must spread
api.getAuthHeader() itself — a missing header is a 401 that only shows up once OIDC is on.
secrets.ts uses raw fetch for historical reasons; it is debt, not a pattern to copy.
UI state
Zustand for chat/debug stores; useState/useCallback everywhere else. No Redux.
i18n — blocking requirement
Every string: t("namespace.key", "English fallback"). Add the key to
src/i18n/locales/en.json, then propagate translations to all 10 other locales
(de, fr, es, ar, zh, th, ja, ko, pt, hi) in the same commit. Each editor gets its own
namespace (llmEditor.*, rulesEditor.*). Because Arabic ships, every layout must use
logical properties.
Tests
- Unit: Vitest + RTL in
__tests__/directories beside the code (pages insrc/pages/__tests__/). Render throughrenderPage(path, element, routePattern?, client?)fromsrc/test/test-utils.tsx— it wrapsMemoryRouter+QueryClient+ThemeProvider. Assert ondata-testid. - Mocks: MSW handlers in
src/test/mocks/handlers.ts. Specific GET handlers must be registered before the genericcreateResourceHandlersblock. Mock data should match the backend Java model. - E2E: Playwright in
e2e/— includingrtl.spec.tsandtheme.spec.ts, so a new screen must survive both Arabic and dark mode.
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.
- 16d ago Changed · +6 lines 8730f5795bfc
- 29d ago First seen · 73 lines · 46 tokens per session scan A 522f3a059d5a
eddi-data is a skill published in the GitHub repository labsai/EDDI (384 stars, last pushed today), licensed Apache-2.0. It adds 46 tokens to every session and 902 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 2026-09-17.
Other skills, from other repositories
aws-cloudformation-lambda
Provides AWS CloudFormation patterns for Lambda functions, layers, API Gateway integration, event sources, cold start optimization, monitoring, logging, template validation, and deployment workflows. Use when creating Lambda functions with CloudFormation, configuring event sources, implementing cold start…
vercel-ai-sdk-streaming
Build streaming AI features in Next.js with Vercel AI SDK, tool calling, RSC/UI hooks, and edge-safe API routes. Use for TypeScript fullstack AI UX with React Server Components.
mongodb-document-modeling
Design MongoDB document schemas, indexes, aggregation pipelines, and multi-tenant patterns with operational safety. Use for document-store backends in fullstack applications.
react-nextjs-frontend
Apply frontend guidance for React and Next.js stacks, including routing, rendering strategy, state boundaries, and performance defaults. Use when a project uses React, Next.js, TypeScript, and component-driven UI delivery.
rules-quarkus
Expert knowledge for building Quarkus applications with Bazel using the rulesquarkus build system. Use when user asks about Quarkus+Bazel builds, augmentation, or troubleshooting.
quarkus-native
Deep expertise in Quarkus Native Image builds, GraalVM integration, reflection configuration, and Profile-Guided Optimization. Use for native compilation questions.