Which integrations are available for this server?

Search for academic papers on arXiv and fetch their PDFs, converting them to Markdown. Search GitHub repositories and fetch repository content or metadata. Search Reddit for posts and comments. Perform privacy-respecting web searches via the SearXNG meta-search engine. Search scientific literature via the Semantic Scholar API. Search Wikipedia articles and fetch their content. Fetch YouTube video transcripts.

How do I use Apeiron?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Apeiron search for latest advancements in AI" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

Apeiron

by insomnia-me

Overview Schema Related Servers Score Discussions

Python

Local

🧬 Philosophy

Why can Anthropic charge $200/month for knowledge, while ordinary people get blocked by Cloudflare?

Information wants to be free. Apeiron is the tool that makes it so.

Every AI agent — OpenCode, Claude Code, Cursor, Cline, any of them — deserves unrestricted access to the open web. Not through paid APIs, not through corporate gatekeepers, but through the collective power of every free anti-bot technology united into one autonomous system.

Apeiron combines:

CloakBrowser (58 C++ patches, 0.9 reCAPTCHA score) — best stealth browser
Markitdown (Microsoft, 147k ⭐) — PDF/DOCX/PPTX → Markdown
yt-dlp (95k ⭐) — YouTube transcripts
SearXNG — privacy-respecting meta-search
Wikipedia / arXiv / Semantic Scholar / Reddit / GitHub APIs
Self-learning — remembers what works per domain, auto-commits new bypass patterns

Apeiron doesn't just scrape. It learns. Every new anti-bot pattern discovered by one user becomes a patch for everyone. The network gets smarter together.

Related MCP server: Forage-mcp

⚡ One-command install

curl -fsSL https://raw.githubusercontent.com/insomnia-me/apeiron/main/install.sh | bash

Then:

# Fetch any URL — bypasses Cloudflare, Turnstile, CAPTCHA, everything
apeiron fetch "https://arxiv.org/pdf/2203.02155.pdf"

# Search across 6 sources at once
apeiron search "quantum computing 2026"

# Teach Apeiron a new domain
apeiron learn "https://protected-site.com"

# Start MCP server for your AI agent
apeiron serve

🔌 MCP server — use from any AI agent

// ~/.config/opencode/opencode.jsonc
{
  "mcp": {
    "servers": {
      "apeiron": {
        "command": "python",
        "args": ["-m", "apeiron.api.mcp_server"]
      }
    }
  }
}

Then your agent calls:

Tool	What it does
`apeiron_search("query")`	Search web + arXiv + Wikipedia + Reddit + GitHub
`apeiron_fetch("url")`	Fetch anything: HTML, PDF, YouTube, bypassing all blocks
`apeiron_learn("url")`	Learn best strategy for a domain

Your agent never writes scraping code. It just asks Apeiron.

🧠 Architecture

                    APEIRON
                        │
          ┌─────────────┴─────────────┐
          │                           │
     SEARCH                      FETCH
          │                           │
  ┌───┬───┼───┬───┬───┐     ┌───────┼──────────┐
  │   │   │   │   │   │     │       │          │
SearXNG  Wiki Reddit  GH  HTTP   BROWSER    MEDIA
  arXiv       pedia          Level  Level     Level
                                   │
                              CloakBrowser
                              Camoufox
                              FlareSolverr
                              browser-use
                                   │
                            ┌──────┴──────┐
                         EXTRACT       SELF-LEARN
                            │              │
                      Trafilatura     strategies.json
                      Markitdown      heuristics.py
                      Readability     git push
                                   (new bypass patterns)

What happens when you call `apeiron.fetch(url)`:

1. Auto-detect content type
   ── YouTube? → yt-dlp → transcript
   ── PDF?     → Markitdown → clean text
   ── HTML?    → go to step 2

2. Check strategy cache
   ── Known domain? → use cached best tier → done
   ── Unknown?      → go to step 3

3. Tier escalation:
   ┌─ 1. curl_cffi (TLS impersonation)        0.2s
   ├─ 2. Patchright (Playwright)              1.5s
   ├─ 3. ★ CloakBrowser (58 C++ patches)      3.0s
   ├─ 4. Camoufox (Firefox C++ patches)       3.0s
   ├─ 5. FlareSolverr (Docker)                5.0s
   ├─ 6. browser-use (AI agent browsing)      15s
   └─ 7. Jina Reader (API fallback)           5.0s

4. Success? → save best tier to strategies.json
   Blocked? → detect challenge pattern → auto-commit to git
              → all users get the update

💻 Python API

from apeiron import search_sync, fetch_sync, learn_sync

# Search across all sources
results = search_sync("transformer architecture 2026")
for r in results:
    print(f"[{r.source}] {r.title}")
    print(f"  {r.url}")
    print(f"  {r.snippet[:100]}\n")

# Fetch any URL — auto-detects content type
content = fetch_sync("https://arxiv.org/pdf/2203.02155.pdf")
# → PDF auto-detected → Markitdown → clean markdown text

content = fetch_sync("https://youtube.com/watch?v=dQw4w9WgXcQ")
# → YouTube auto-detected → yt-dlp → transcript

content = fetch_sync("https://cloudflare-protected-site.com")
# → HTML auto-detected → tier escalation → CloakBrowser → text

# Teach a new domain
result = learn_sync("https://example-protected-site.com")
print(f"Best tier: {result.tier}")  # → cloakbrowser

⚔️ vs alternatives

Feature	Firecrawl	Crawl4AI	browser-use.com	Apeiron
Stealth browser	❌	❌	💰 $49-299/mo	✅ CloakBrowser (free)
PDF → Markdown	❌	❌	❌	✅ Markitdown (147k ⭐)
YouTube transcripts	❌	❌	❌	✅ yt-dlp (95k ⭐)
Multi-source search	❌	❌	❌	✅ SearXNG + 5 APIs
Self-learning	❌	❌	❌	✅ strategies.json
Auto-commit bypasses	❌	❌	❌	✅ git push to all
MCP server	❌	❌	❌	✅ MCP protocol
Price	Free tier	Free	$49-599/mo	Free forever

📦 Install

Quick (recommended)

curl -fsSL https://raw.githubusercontent.com/insomnia-me/apeiron/main/install.sh | bash

pip

pip install apeiron
# With all features:
pip install "apeiron[all]"

From source

git clone https://github.com/insomnia-me/apeiron.git
cd apeiron
python3 -m venv .venv
source .venv/bin/activate
pip install -e ".[all]"

Docker infrastructure (optional — for SearXNG + FlareSolverr)

bash scripts/start-infra.sh

🧪 Commands

apeiron search "query"                    # Multi-source search
apeiron fetch "https://..."               # Fetch any URL
apeiron fetch "https://..." -o output.md  # Save to file
apeiron learn "https://..."               # Train on new domain
apeiron serve                             # Start MCP server

🏛️ How self-learning works

User fetches URL
       ↓
Tier 1 fails (blocked by Cloudflare)
       ↓
Tier 2 fails (Turnstile challenge)
       ↓
Tier 3 succeeds (CloakBrowser bypasses)
       ↓
✓ Save "cloakbrowser" as best tier for this domain
  in strategies.json
       ↓
✓ If new anti-bot pattern detected:
  → Add to heuristics.py
  → Git commit & push
  → All Apeiron users get the update
       ↓
Next time: direct to CloakBrowser for this domain

The network grows smarter with every fetch.

🛠️ What's under the hood

Category	Technology	Stars	License
Stealth browser	CloakBrowser	25k ⭐	MIT
Document conversion	Markitdown (Microsoft)	147k ⭐	MIT
YouTube transcripts	yt-dlp	95k ⭐	Unlicense
Meta-search	SearXNG	—	AGPL
TLS impersonation	curl_cffi	3k ⭐	MIT
Firefox stealth	Camoufox	5k ⭐	MIT
AI browsing	browser-use	70k ⭐	MIT
AI extraction	ScrapeGraphAI	25k ⭐	MIT
Content extraction	Trafilatura	3k ⭐	GPL
Article extraction	Readability (Mozilla)	8k ⭐	Apache
Scientific search	Semantic Scholar API	—	Free
Knowledge base	Wikipedia API	—	Free

This server cannot be installed

license - permissive license

quality - not tested

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

–Release cycle

1Releases (12mo)

Commit activity

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

Lightport: Open-Sourcing Glama's AI Gateway
By punkpeye on April 27, 2026.
open source
OpenAI
Tool Definition Quality Score (TDQS)
By punkpeye on April 3, 2026.
mcp
The Hackers Who Tracked My Sleep Cycle
By punkpeye on March 26, 2026.
security

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/insomnia-me/apeiron'

If you have feedback or need assistance with the MCP directory API, please join our Discord server