request-smuggling-specialist

Triages HTTP request smuggling and desync signals into structured variant hypotheses.

7|4|Updated Jun 22, 2026
One-click install
npx skills add https://github.com/dbx0/skills --skill request-smuggling-specialist-dbx0
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: request-smuggling-specialist
Source: https://github.com/dbx0/skills/tree/main/skills/web-appsec/initial-access/request-smuggling-specialist
Command: npx skills add https://github.com/dbx0/skills --skill request-smuggling-specialist-dbx0

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? HTTP request smuggling and desync signals from scanners are noisy and hard to act on. This Skill routes a broad smuggling signal through a triage router, selects at most three relevant technique cards from 32 variants (CL.TE, TE.CL, H2.CL, CL.0, h2c, HTTP/3 contamination, and more), and produces a structured assessment with explicit false-positive controls and safety gates. ## Core Features & Use Cases - Signal Routing: A router maps evidence types (conflicting CL/TE, HTTP/2 downgrade, early response, cache anomalies) to specific technique cards instead of loading everything. - 32 Technique Cards: Each card defines when to consider the variant, evidence to collect, false-positive controls, and safe validation boundaries. - Structured Output: Enforces an output schema with status values like needs_more_evidence, hypothesis_ready, and manual_proof_contract_required. - Safety Gates: Separates an automatic lane (GET/HEAD inventory, passive observation) from a manual-gated lane (malformed framing, queue poisoning, timing probes). - Use Case: A scanner flags possible CL.TE desync on an API gateway. The Skill routes the signal, loads the cl-te and te-cl cards, and returns a hypothesis naming the parser boundary, missing evidence, and one safe next step. ## Quick Start Ask the agent to triage a suspected HTTP request smuggling signal on a target host and return a structured desync assessment using the router and output schema.

Frequently Asked Questions about request-smuggling-specialist

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I triage a suspected HTTP request smuggling finding?

Start with the router: confirm scope, host, and evidence, classify the signal type, then load at most three technique cards matching the routing hints. Return structured output with status, variant hypotheses, parser boundary, and a safe next step.

What request smuggling variants does this skill cover?

It covers 32 technique cards including CL.TE, TE.CL, TE.TE, CL.0, 0.CL, H2.CL, H2.TE, H2.0, HTTP/2 request splitting, request tunnelling, response queue poisoning, client-side and browser-powered desync, h2c upgrade smuggling, HTTP/3 connection contamination, and TRACE-assisted desync.

Does this skill automatically exploit request smuggling vulnerabilities?

No. It is a triage specialist, not an exploit pack. Automatic work is limited to GET/HEAD inventory and passive observation; malformed framing, timing probes, and cache or queue poisoning all require manual approval under its safety gates.

How does the skill handle scanner-only desync findings?

Scanner-only signals without raw request/response metadata or architecture evidence return a needs_more_evidence status. The router asks for specific missing evidence rather than guessing a variant.

What are the limitations of this desync triage approach?

It produces hypotheses, not confirmed findings, and never validates actively without manual approval. It cannot distinguish all false positives such as HTTP pipelining artifacts without connection-level evidence collected manually.