Bypass WAF payload inspection using HTTP/2 binary framing attacks.
npx skills add https://github.com/dreadnode/capabilities --skill h2-waf-bypass
Please help me install this Agent Skill. Skill: h2-waf-bypass Source: https://github.com/dreadnode/capabilities/tree/main/capabilities/web-security/skills/h2-waf-bypass Command: npx skills add https://github.com/dreadnode/capabilities --skill h2-waf-bypass
Standard web application firewalls (WAFs) often block malicious payloads sent over HTTP/1.1, but HTTP/2's binary framing architecture creates blind spots for out-of-process WAFs that evaluate requests before the full request body is received. This Skill enables bypassing these WAF restrictions when standard 403 bypass and parser differential techniques fail, and the target supports HTTP/2.
Use the h2-waf-bypass skill to run the full WAF bypass test pipeline against your target host and port to identify exploitable HTTP/2 framing vulnerabilities.
High-intent search queries and answers about installing and using this skill.
WAF bypass can be achieved by exploiting HTTP/2 binary framing blind spots in out-of-process WAF architectures, which evaluates payloads differently than the backend server. This technique applies when the target supports HTTP/2 and standard parser differential methods have been exhausted.
HTTP/2 framing WAF evasion is generally ineffective against hardened configurations like nginx with libmodsecurity3. Automated fingerprinting identifies the reverse proxy and WAF architecture beforehand to avoid wasting effort on these hardened setups.
For authorized penetration testing, you can use six distinct HTTP/2 framing bypasses including body timing attacks, body size truncation, Extended CONNECT method conversion, path normalization bypasses, JSON content-type gap exploitation, and ForwardAuth body stripping.
Proxy fingerprinting identifies the specific reverse proxy and WAF architecture in use, allowing selective execution of context-aware HTTP/2 framing attack vectors. This prevents wasted effort on configurations that are not vulnerable to specific bypass techniques.
Out-of-process WAFs miss HTTP/2 payloads because their binary framing architecture creates blind spots when evaluating requests before the full body is received. This parser differential allows malicious content to pass inspection while being interpreted differently by the backend.
No external Python libraries are required to run HTTP/2 WAF bypass tests. The automated fingerprinting, selective attack vector execution, and evidence capture for confirmed bypasses operate using only Python standard library dependencies.
DorianGalloIdentifies web application firewalls and selects generic evasion techniques for injection payloads.
dreadnodeExtract database data from blind SQL injection points using LIKE and DIV methods.
Prohao42Bypass 401/403 access controls using path, method, header, and protocol techniques.
ok-helloworldIdentify and test HTTP/2-specific request-smuggling and desynchronization vulnerabilities.
