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 sebaskauf/agentic-os --skill tiefe-recherchegit clone --depth 1 https://github.com/sebaskauf/agentic-osWrote 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/sebaskauf/agentic-os/tiefe-recherche)<a href="https://agentmods.dev/skills/sebaskauf/agentic-os/tiefe-recherche"><img src="https://agentmods.dev/badge/skills/sebaskauf/agentic-os/tiefe-recherche/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/sebaskauf/agentic-os/tiefe-recherche"><img src="https://agentmods.dev/badge/skills/sebaskauf/agentic-os/tiefe-recherche.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 | $0.00113 | $0.01237 |
| Opus 5 | $0.00056 | $0.00619 |
| Sonnet 5 | $0.00023 | $0.00247 |
| Haiku 4.5 | $0.00011 | $0.00124 |
Grade A, and why
tiefe-recherche 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 9d 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 — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tiefe Recherche
Dein Job: eine Frage nicht oberflaechlich beantworten, sondern wie ein skeptischer Researcher. Mehrere Quellen lesen, Aussagen gegeneinander pruefen, Widersprueche benennen, alles belegen. Lieber "das ist unklar" sagen als raten.
Tools die du nutzt: WebSearch (Quellen finden), WebFetch (Quellen lesen), Read (lokale Dateien falls relevant), Write (nur am Ende fuer die optionale Notiz). Bis zum Speichern ist alles read-only.
0. Frage schaerfen (10 Sekunden)
Wenn die Frage zu vage ist, um sinnvoll zu recherchieren, stell 1-2 kurze Rueckfragen. Beispiele fuer zu vage:
- "recherchier mal E-Autos" → Welches Ziel? Kaufberatung, Markttrend, Technik-Vergleich?
- "finde raus was besser ist" → Besser wofuer, fuer wen, mit welchem Budget?
Wenn die Frage klar genug ist, leg direkt los. Nicht unnoetig nachfragen.
1. Quellen finden (Fan-out)
- Zerlege die Frage in 2-4 Such-Anfragen aus verschiedenen Winkeln. Bei "X vs Y" je eine Suche pro Seite plus eine neutrale Vergleichssuche.
- Fuehre die
WebSearch-Anfragen aus. Bei aktuellen Themen Datum/Jahr in die Query packen. - Waehle die vielversprechendsten Treffer. Mindestens 3 voneinander unabhaengige Quellen (nicht 3x dieselbe Pressemitteilung in anderem Gewand).
- Bevorzuge Primaerquellen, offizielle Doku, Studien, etablierte Medien. SEO-Spam, reine Werbe-Landingpages und Inhalte ohne Autor/Datum niedriger gewichten.
2. Quellen lesen und querchecken
- Lies jede ausgewaehlte Quelle mit
WebFetch. Pro Quelle festhalten: Kernaussage, Datum, wer steht dahinter, wie vertrauenswuerdig. - Triangulieren: Bestaetigen sich die Aussagen ueber Quellen hinweg? Was sagen nur einzelne Quellen?
- Widersprueche markieren: Wenn zwei Quellen sich widersprechen, beide nennen und sagen welche glaubwuerdiger ist und warum, statt eine still zu schlucken.
- Skepsis: Achte auf Interessenkonflikte (Hersteller lobt eigenes Produkt), veraltete Daten, fehlende Belege, Behauptungen ohne Quelle. Eine einzelne unbelegte Behauptung ist kein Fakt.
- Wenn die 3 Quellen nicht reichen oder uneinig sind, noch eine Runde suchen statt zu raten.
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.
- 9d ago First seen · 93 lines · 113 tokens per session scan A 55692a346323
tiefe-recherche is a skill published in the GitHub repository sebaskauf/agentic-os (2 stars, last pushed 1mo ago), licensed MIT. It adds 113 tokens to every session and 1,237 once invoked, about $0.0006 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.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
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…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
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…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…