agentmods.dev

deep-research

deep-research is a skill for Claude Code from yachi/claude-skills. It costs 164 tokens per session (3,821 once invoked), scanned A, original, MIT.

A research mode for investigating a topic using verified evidence from sources such as technical documents, studies, and official records.

In plain words
What is it for?
Use it for thorough technical, business, scientific, regulatory, financial, medical, or legal investigations that need evidence and reproducible findings.
Why use it?
It addresses the problem of making important claims without checking them or testing whether opposing evidence exists.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: positional $N argument.

Part of the deep-research plugin — 1 skill shipped together

not rated 3repo 5mo ago A scan 164 tokens original MIT

Good fit Use it for thorough technical, business, scientific, regulatory, financial, medical, or legal investigations that need evidence and reproducible findings.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/yachi/claude-skills/deep-research
View source ↗ yachi/claude-skills
Install

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.

Any agent
npx skills add yachi/claude-skills --skill deep-research
Clone the repo
git clone --depth 1 https://github.com/yachi/claude-skills

Made for: Claude Code.

Or install deep-research, the plugin that ships this one along with the rest of its 1 skill.

Wrote 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.

agentmods badge for deep-research

README.md
[![agentmods](https://agentmods.dev/badge/skills/yachi/claude-skills/deep-research/github.svg)](https://agentmods.dev/skills/yachi/claude-skills/deep-research)
Your own site
<a href="https://agentmods.dev/skills/yachi/claude-skills/deep-research"><img src="https://agentmods.dev/badge/skills/yachi/claude-skills/deep-research/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.

agentmods 80×15 button for deep-research

Your own site · 80×15
<a href="https://agentmods.dev/skills/yachi/claude-skills/deep-research"><img src="https://agentmods.dev/badge/skills/yachi/claude-skills/deep-research.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 164 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,821 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00164 $0.03821
Opus 5 $0.00082 $0.01911
Sonnet 5 $0.00033 $0.00764
Haiku 4.5 $0.00016 $0.00382

Measured 10d ago against content hash 72d87fe40118, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

deep-research 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 10d 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.

skills/deep-research/SKILL.md · 201 lines

How it starts

The opening of the file, as written. The whole thing — 201 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/dr — Deep Research

You are now in Deep Research mode. The standard: ICD 203 analytic tradecraft (9/9 standards), GRADE evidence hierarchy, NIST reproducibility. Every claim survives cross-examination by the foremost domain expert. No hand-waving. No unverified assertions. If you can't prove it, you don't say it.

Core Principles

  1. Evidence over reasoning — Your internal knowledge is a starting hypothesis, never a conclusion. Every factual claim must be verified through tool use (WebSearch, WebFetch, Bash, code execution, MCP tools) against authoritative external sources.

  2. Quantitative over qualitative — Wherever possible, replace adjectives with numbers. "Faster" becomes "47% lower p99 latency (benchmarked)." "More popular" becomes "3.2M weekly downloads vs 180K." "More secure" becomes "0 CVEs in 3 years vs 12 (4 critical)."

  3. Standards over opinions — Every domain has authoritative standards bodies, best practices, and regulatory frameworks. Find them. Cite them. Compare the subject against them.

  4. Adversarial rigor — Before presenting conclusions, actively try to disprove them. The strongest argument is one that has survived its best counterargument.

Research Protocol

Phase 0: Clarify (if needed)

If the query is ambiguous or underspecified, ask 1-2 clarifying questions before committing to research. Skip if the question is already clear and specific.

Phase 1: Decompose

Break the research question into orthogonal sub-questions. For each sub-question, identify:

  • What specifically needs to be answered
  • What type of evidence would be conclusive (benchmark data, standards compliance, expert consensus, regulatory text, academic study)
  • Where that evidence is most likely found

Do this decomposition explicitly — write it out so the user can see your research plan.

Premise validation (kill switch): Identify the core premise. Quick-search to verify it holds before investing in all phases. If it doesn't, pivot immediately and tell the user. A fast pivot based on evidence saves enormous wasted effort and is more valuable than a thorough analysis built on a false premise.

Read the full file on GitHub · 201 lines

Files

What ships with it

60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

  1. 10d ago First seen · 201 lines · 164 tokens per session scan A 72d87fe40118

Subscribe to this mod's changes

deep-research is a skill published in the GitHub repository yachi/claude-skills (3 stars, last pushed 5mo ago), licensed MIT. It adds 164 tokens to every session and 3,821 once invoked, about $0.0008 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.

Claude CodeResearchai-skillsai-toolsanthropicclaudeclaude-code-plugin
Related

Other skills, from other repositories

cco-tools

Show what tools actually cost in tokens — learned per-tool averages from observed results, replacing the hardcoded MCP/Agent guesses.

egorfedorov/claude-context-optimizer · 27 tokens

cco-clean

Clean up old tracking data and reset statistics.

egorfedorov/claude-context-optimizer · 11 tokens

skill-creator-plus

Create, test, evaluate, and improve Claude skills — and answer questions about how skills work. Use for "create a skill", "make a skill for", "write a SKILL.md", "turn this into a skill", "run evals", "test/benchmark my skill", "optimize my skill description", "improve triggering", "A/B test my skill", "blind…

yaniv-golan/skill-creator-plus · 147 tokens

openclaw-setup

Set up, install, configure, and deploy OpenClaw (formerly ClawdBot/MoltBot) — a personal AI assistant that runs on your own devices and connects to messaging channels. Use when users ask to "set up OpenClaw," "install ClawdBot," "install MoltBot," "deploy a personal AI assistant," "configure OpenClaw on Mac," "deploy…

Aznatkoiny/zAI-Skills · 188 tokens

cpp-reinforcement-learning

C++ Reinforcement Learning best practices using libtorch (PyTorch C++ frontend) and modern C++17/20. Use when: Implementing RL algorithms in C++ for performance-critical applications Building production RL systems with libtorch Creating replay buffers and experience storage Optimizing RL training with GPU acceleration…

Aznatkoiny/zAI-Skills · 83 tokens

fp-yoohoo

Terminal-recall mode. Reference for fp-yoohoo, the plugin that brings the terminal window to the foreground when Claude Code stops and waits on a human: a permission prompt, a question dialog, a plan to approve. Explains what it hooks, which desktops it can raise a window on, how the already-focused guard works, and…

SlashyIsTaken/fp-pluginplaza · 161 tokens