New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Deep Searcher Guide

by wentorai in wentorai/research-plugins

Open deep research alternative for private data with vector search

MITAuto-check passedResearch & Science

Install Deep Searcher Guide

skills CLI
$ npx skills add wentorai/research-plugins --skill deep-searcher-guide -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install wentorai/research-plugins deep-searcher-guide --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/wentorai/research-plugins.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/research/deep-research/deep-searcher-guide .claude/skills/deep-searcher-guide && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
deep-searcher-guide
GitHub stars
298
Used in
1 other repo
Token cost
~2.1k tokens
SKILL.md length
532 words
Files
1
Skills in repo
405
Repo updated
First seen
Licence
MIT

At a glance

Open deep research alternative for private data with vector search

  • Works in 5 steps: Query decomposition: The research… → Initial retrieval: Vector search… → Analysis: The LLM analyzes retrieved… → …
  • Tasks that involve Deep research
  • SKILL.md covers Overview, Installation and Setup, Document Ingestion and Deep Research Workflow, plus 3 more sections
  • Calls pip, git and docker; reaches github.com; needs OPENAI_API_KEY and TAVILY_API_KEY

What it does

Deep Searcher Guide is an agent skill from wentorai/research-plugins. Open deep research alternative for private data with vector search

Its SKILL.md is about 2.1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Research & Science, covering Deep research and Vector databases. It works with Milvus. The repository describes itself as: 350+ academic research skills, MCP configs, and plugins for Research-Claw and AI agents. The licence is MIT.

When your agent uses it

  • Tasks that involve Deep research
  • Tasks that involve Vector databases

Example prompts

  • “/deep-searcher-guide”

Requirements

  • Python 3
  • Docker
  • A credential in OPENAI_API_KEY
  • A credential in TAVILY_API_KEY

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Query decomposition: The research question is broken into sub-queries
  2. Initial retrieval: Vector search retrieves relevant passages for each sub-query
  3. Analysis: The LLM analyzes retrieved content and identifies information gaps
  4. Refined search: New queries are generated to fill gaps, with the search refined based on what has been found
  5. Synthesis: All gathered information is synthesized into a comprehensive answer with citations

What it can do on your machine

Read from SKILL.md and the folder at commit bf44b3c. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • pip
    • git
    • docker
    • curl

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • github.com

    Also links to:

    • milvus.io
    • zilliz.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • OPENAI_API_KEY
    • TAVILY_API_KEY

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Deep Searcher Guide loads about 2.1k tokens when it runs. Until then it costs about 22 tokens; SKILL.md has 532 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~22
When it runs · the whole SKILL.md, loaded when a task matches
~2.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from wentorai/research-plugins at commit bf44b3c, republished under its MIT licence (© wentorai). 532 words, ~2,081 tokens.

Download SKILL.mdSave it as .claude/skills/deep-searcher-guide/SKILL.md (or your agent's skills folder).
name
deep-searcher-guide
description
Open deep research alternative for private data with vector search

Deep Searcher Guide

Overview

Deep Searcher is an open-source deep research tool developed by Zilliz with over 8,000 GitHub stars, designed to be an open alternative to proprietary deep research systems like OpenAI's Deep Research and Gemini Deep Research. What distinguishes Deep Searcher is its focus on private data -- it enables researchers to conduct deep, iterative research over their own document collections, databases, and institutional knowledge bases rather than being limited to public web content.

The system combines vector search via Milvus (or other vector databases) with agentic RAG (Retrieval-Augmented Generation) to decompose complex research questions, retrieve relevant passages from your document collection, reason over the retrieved content, and iteratively refine its search until it can produce a comprehensive answer. This makes it particularly valuable for researchers who work with proprietary datasets, unpublished manuscripts, internal reports, or specialized domain corpora that are not available through web search.

Deep Searcher supports multiple LLM providers and embedding models, and can be deployed entirely on-premises for organizations with strict data privacy requirements. It is built on top of Milvus, the high-performance open-source vector database also created by Zilliz, ensuring scalable and efficient similarity search across large document collections.

Installation and Setup

bash
# Install Deep Searcher
pip install deepsearcher

# Or clone for development
git clone https://github.com/zilliztech/deep-searcher.git
cd deep-searcher
pip install -e .
Dependencies Setup

Deep Searcher requires a vector database and LLM access:

bash
# Option 1: Milvus Lite (embedded, no separate server needed)
pip install pymilvus[model]

# Option 2: Full Milvus via Docker
docker run -d --name milvus \
  -p 19530:19530 \
  -p 9091:9091 \
  milvusio/milvus:latest standalone

# Configure LLM access
export OPENAI_API_KEY=$OPENAI_API_KEY
# Or for local LLMs
export OLLAMA_BASE_URL=http://localhost:11434
Configuration

Create a configuration file for your research setup:

python
from deepsearcher import DeepSearcher
from deepsearcher.config import Config

config = Config(
    # Vector database settings
    vector_db="milvus_lite",  # or "milvus", "zilliz_cloud"
    collection_name="research_papers",

    # LLM settings
    llm_provider="openai",
    llm_model="gpt-4o",

    # Embedding settings
    embedding_model="text-embedding-3-small",

    # Research settings
    max_iterations=10,
    chunk_size=1000,
    chunk_overlap=200,
)

searcher = DeepSearcher(config)

Document Ingestion

Loading Research Documents

Ingest your research documents into the vector database for searchable access:

python
# Load individual files
searcher.load_document("path/to/paper.pdf")
searcher.load_document("path/to/notes.md")

# Load entire directories
searcher.load_directory(
    "path/to/papers/",
    file_types=["pdf", "md", "txt", "docx"],
    recursive=True,
)

# Load with metadata for filtering
searcher.load_document(
    "path/to/paper.pdf",
    metadata={
        "author": "Smith et al.",
        "year": 2024,
        "topic": "transformer efficiency",
        "venue": "NeurIPS",
    }
)
Supported Document Types

Deep Searcher supports a wide range of document formats commonly used in academic research:

  • PDF: Research papers, textbooks, reports (with OCR support for scanned documents)
  • Markdown: Research notes, documentation, wikis
  • Plain text: Data files, logs, transcripts
  • DOCX/DOC: Word documents, manuscripts
  • HTML: Web pages, saved articles
  • LaTeX: TeX source files with equation extraction
  • Jupyter Notebooks: Code and analysis notebooks

Deep Research Workflow

Basic Research Query
python
# Ask a research question over your document collection
result = searcher.research(
    query="What methods have been proposed for reducing the "
          "computational complexity of self-attention in transformers?",
)

print(result.answer)
print(f"Sources: {len(result.sources)}")
for source in result.sources:
    print(f"  - {source.document}: {source.chunk_preview[:100]}...")
Show full SKILL.md (231 more words)Show less
Iterative Research Process

Deep Searcher follows an iterative research pipeline:

  1. Query decomposition: The research question is broken into sub-queries
  2. Initial retrieval: Vector search retrieves relevant passages for each sub-query
  3. Analysis: The LLM analyzes retrieved content and identifies information gaps
  4. Refined search: New queries are generated to fill gaps, with the search refined based on what has been found
  5. Synthesis: All gathered information is synthesized into a comprehensive answer with citations
python
# Watch the iterative research process
result = searcher.research(
    query="Compare the approaches to efficient attention in the papers "
          "I have collected, focusing on trade-offs between speed and quality",
    verbose=True,  # Print each research iteration
    max_iterations=8,
)

# Access the research trace
for step in result.trace:
    print(f"Iteration {step.iteration}:")
    print(f"  Sub-query: {step.query}")
    print(f"  Documents found: {step.num_results}")
    print(f"  Gap identified: {step.gap}")
Filtered Research

Narrow your research to specific subsets of your collection:

python
# Research only within papers from a specific venue
result = searcher.research(
    query="Novel loss functions for contrastive learning",
    filters={"venue": "ICML", "year": {"$gte": 2023}},
)

# Research across specific document groups
result = searcher.research(
    query="How do the baseline methods compare across my experiment logs?",
    filters={"topic": "baseline-comparison"},
)

Integration with Research Tools

Combining Private and Public Data

Deep Searcher can be combined with web search for comprehensive research that covers both your private collection and public sources:

python
from deepsearcher.sources import WebSearchSource

# Add web search as an additional source
config.add_source(WebSearchSource(
    provider="tavily",
    api_key_env="TAVILY_API_KEY",
))

# Research now spans both private documents and the web
result = searcher.research(
    query="Recent advances in protein folding prediction",
    sources=["private", "web"],
)
Export and Sharing

Export research results in formats suitable for academic use:

python
# Export as markdown report
result.export_markdown("research_report.md")

# Export citations in BibTeX format
result.export_citations("references.bib")

# Export the full research trace for reproducibility
result.export_trace("research_trace.json")
API Server

Run Deep Searcher as a service for team-wide access:

bash
# Start the API server
deepsearcher serve --host 0.0.0.0 --port 8000

# Query via REST API
curl -X POST http://localhost:8000/research \
  -H "Content-Type: application/json" \
  -d '{"query": "What are the key findings in our latest experiments?"}'

Performance and Scalability

Deep Searcher leverages Milvus for high-performance vector search, which means it can handle document collections ranging from hundreds to millions of documents efficiently. Key performance considerations include:

  • Indexing: Milvus uses HNSW or IVF indexes for fast approximate nearest neighbor search
  • Chunking strategy: Adjustable chunk size and overlap to balance retrieval precision and recall
  • Embedding caching: Previously computed embeddings are cached to avoid redundant computation
  • Batch processing: Documents can be ingested in parallel for faster indexing

References

© wentorai, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/research/deep-research/deep-searcher-guide of wentorai/research-plugins.

Open the folder on GitHubat commit bf44b3c

Used in 1 other repository

We found 1 copy of this SKILL.md (exact, near-identical or edited) in other folders, from 1 other GitHub owner. This page covers the copy in wentorai/research-plugins, which our catalogue first saw on October 7, 2026.

Compare with similar skills

Deep Searcher Guide next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Deep Searcher Guide compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Deep Searcher Guide this skillwentorai/research-plugins2981 repos~2.1kAutomated safety check: PassMIT
GitHub Deep Researchbytedance/deer-flow84k4 repos~1.3kAutomated safety check: PassMIT
Deep Research WorkflowTokenRhythm/opensquilla7.1k—~1.3kAutomated safety check: PassApache-2.0
Deep Researchsanjay3290/ai-skills4329 repos~683Automated safety check: NotesApache-2.0
Horizontal-Vertical Deep ResearchKKKKhazix/khazix-skills21k—~2.1kAutomated safety check: PassMIT
Academic Research PipelineImbad0202/academic-research-skills51k—~15kAutomated safety check: PassCustom licence

Similar skills

  • GitHub Deep Research

    bytedance/deer-flow

    Researches a GitHub repository over four rounds using the GitHub API and web search, then writes a structured markdown report with timeline, metrics and Mermaid diagrams.

    84k GitHub starsUsed in 4 repos~1.3k tokens
    Research & ScienceAuto-check passed

  • Deep Research Workflow

    TokenRhythm/opensquilla

    Runs multi-round research in three stages with a persisted state file, evidence tracking and a long-form report with per-claim citations.

    7.1k GitHub stars~1.3k tokensUpdated today
    Research & ScienceAuto-check passed

  • Deep Research

    sanjay3290/ai-skills

    Execute autonomous multi-step research using Google Gemini Deep Research Agent.

    432 GitHub starsUsed in 9 repos~683 tokens
    Research & ScienceAuto-check: notes

  • Horizontal-Vertical Deep Research

    KKKKhazix/khazix-skills

    Runs a two-axis deep research method on a product, company, concept or person: its full history over time, compared with peers today, delivered as a typeset PDF report.

    21k GitHub stars~2.1k tokensUpdated 8 days ago
    Research & ScienceAuto-check passed

  • Academic Research Pipeline

    Imbad0202/academic-research-skills

    Orchestrates a ten-stage academic workflow from research to finished manuscript, including integrity checks, two rounds of peer review and revision.

    51k GitHub stars~15k tokensUpdated yesterday
    Research & ScienceAuto-check passed

  • Cognee Community Packages

    topoteretes/cognee

    Guide to using and contributing cognee community packages: database adapters, data-source connectors, custom tasks and retrievers, and Keywords AI observability.

    32k GitHub stars~1.2k tokensUpdated today
    DatabasesAuto-check passed

More from wentorai/research-plugins

All 405 skills in this repo

  • Abstract Writing Guide

    wentorai/research-plugins

    Craft structured research abstracts that maximize clarity and journal acceptance

    298 GitHub starsUsed in 1 repo~1.7k tokens
    Auto-check passed

  • Academic Citation Manager

    wentorai/research-plugins

    Manage academic citations across BibTeX, APA, MLA, and Chicago formats

    298 GitHub starsUsed in 1 repo~2.7k tokens
    Auto-check passed

  • Academic Paper Summarizer

    wentorai/research-plugins

    Summarize academic papers with structured extraction of key elements

    298 GitHub starsUsed in 1 repo~1.4k tokens
    Auto-check passed

  • Academic Study Methods

    wentorai/research-plugins

    Evidence-based study techniques for academic learning and retention

    298 GitHub starsUsed in 1 repo~1.8k tokens
    Auto-check passed

  • Academic Tone Guide

    wentorai/research-plugins

    Adjust writing tone and register for academic audiences and venues

    298 GitHub starsUsed in 1 repo~1.9k tokens
    Auto-check passed

  • Academic Translation Guide

    wentorai/research-plugins

    Academic translation, post-editing, and Chinglish correction guide

    298 GitHub starsUsed in 1 repo~1.6k tokens
    Auto-check passed

Works with

Categories

Questions about Deep Searcher Guide

What does Deep Searcher Guide do?

Open deep research alternative for private data with vector search. Deep Searcher Guide is an agent skill from wentorai/research-plugins.

When should I use Deep Searcher Guide?

Deep Searcher Guide fits situations like: tasks that involve Deep research; tasks that involve Vector databases.

How do I install Deep Searcher Guide in Claude Code?

Run `npx skills add wentorai/research-plugins --skill deep-searcher-guide -a claude-code`. Or copy the skill folder (skills/research/deep-research/deep-searcher-guide in wentorai/research-plugins) into .claude/skills/deep-searcher-guide in your project. Claude Code loads it when a task matches its description.

How do I install Deep Searcher Guide in Codex?

Run `npx skills add wentorai/research-plugins --skill deep-searcher-guide -a codex`. Or copy the skill folder (skills/research/deep-research/deep-searcher-guide in wentorai/research-plugins) into .agents/skills/deep-searcher-guide in your project. Codex loads it when a task matches its description.

Can I use Deep Searcher Guide in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add wentorai/research-plugins --skill deep-searcher-guide -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/deep-searcher-guide, .gemini/skills/deep-searcher-guide, .github/skills/deep-searcher-guide and .opencode/skills/deep-searcher-guide in your project.

What does Deep Searcher Guide need to run?

Going by SKILL.md and its folder, Deep Searcher Guide needs the command-line tools its instructions call (pip, git, docker and curl) and credentials named OPENAI_API_KEY and TAVILY_API_KEY. Our summary lists: Python 3; Docker; A credential in OPENAI_API_KEY; A credential in TAVILY_API_KEY.

Does Deep Searcher Guide access the network?

SKILL.md names 3 domains. In commands or code: github.com; the agent is likely to contact it when it follows the instructions. As links in the text: milvus.io and zilliz.com. This is read from the text; nothing was executed.

Is Deep Searcher Guide safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Deep Searcher Guide use?

Deep Searcher Guide is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Deep Searcher Guide use?

About 2.1k tokens (SKILL.md is roughly 8.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Deep Searcher Guide?

Skills that share tags, products or a category with Deep Searcher Guide: GitHub Deep Research (bytedance/deer-flow, 84k stars), Deep Research Workflow (TokenRhythm/opensquilla, 7.1k stars), Deep Research (sanjay3290/ai-skills, 432 stars) and Horizontal-Vertical Deep Research (KKKKhazix/khazix-skills, 21k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Deep Searcher Guide?

wentorai (a GitHub user) maintains it in wentorai/research-plugins, which has 298 GitHub stars. The repository holds 405 skills in this directory. The repository was last updated on June 19, 2026.

Source: wentorai/research-plugins on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.