hunting-http-request-smuggling-and-desync SKILL.md skillA

hunting-http-request-smuggling-and-desync from unboundcompute/security-agent-skills is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add unboundcompute/security-agent-skills/hunting-http-request-smuggling-and-desync@git:20260829.b60f9b2

Pins git:20260829.b60f9b2 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
unboundcompute/security-agent-skills
Path
skills/hunting-http-request-smuggling-and-desync/SKILL.md
Versions recorded
1
Latest version
git:20260829.b60f9b2, committed Aug 29, 2026
Size
9,600 bytes
Repository stars
26
License
MIT

Hunt for request smuggling where two HTTP processors on the same path disagree about where one request ends and the next begins: a front end and back end that resolve conflicting Content-Length and Transfer-Encoding headers differently, a proxy that forwards a body the origin re-parses, a keep-alive connection where a smuggled prefix poisons the next user's request, and a rewrite or normalization difference that desyncs the stream. Covers chained HTTP/1.1 processors, proxies, load balancers…

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# Hunting HTTP request smuggling and desync: when two parsers disagree on where a request ends

Request smuggling is a disagreement bug, not a parsing bug in isolation. When a request passes through more
than one HTTP processor, a proxy or load balancer in front of an origin, each one has to decide where one
request ends and the next begins, and if they decide differently, an attacker can hide a second request
inside the first. The classic lever is conflicting length signals: a Content-Length and a Transfer-Encoding
header, or two of either, that the front end and back end resolve in opposite orders. The consequence is
severe because these connections are reused: a smuggled prefix left on a keep-alive connection prepends to
the next user's request, so it poisons other people's traffic, bypasses front-end access controls, and
captures requests. The hunt is about the seam between two parsers, so you look for where their framing rules
differ. You hunt this by finding chained processors and testing whether a crafted request desyncs them.

## When to use

- A request passes through more than one HTTP/1.1 processor: a proxy or load balancer in front of an origin.
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add unboundcompute/security-agent-skills/hunting-http-request-smuggling-and-desync@sha256:e3aa5824c8e8f726

The line at the top of this page pins the label git:20260829.b60f9b2, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_agdd455qvse6sumi.svg)](https://markdownregistry.com/a/art_agdd455qvse6sumi)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 1 version of this file, committed on Aug 29, 2026.

versioncommittedcommitsizeaudit
git:20260829.b60f9b2 latest2026-08-29 b60f9b2 9,600 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (9600 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

unboundcompute/security-agent-skills · 26 stars · license MIT · pushed 2026-09-08 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_agdd455qvse6sumi
GET https://markdownregistry.com/api/v1/resolve?ref=unboundcompute/security-agent-skills/hunting-http-request-smuggling-and-desync
GET https://markdownregistry.com/api/v1/blob/e3aa5824c8e8f7267bc8f62ef81ebcc1121e145b3f3c47826699dfa26e212986

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from unboundcompute/security-agent-skills

adjudicating-dependency-cve-reachability SKILL.md skill
unboundcompute/security-agent-skills · skills/adjudicating-dependency-cve-reachability/SKILL.md · Decide whether a CVE in a dependency actually exposes your application before you scramble to patch: is the vulnerable…
git:20260816.c75b58d · audit A · 26 stars
adjudicating-taint-paths SKILL.md skill
unboundcompute/security-agent-skills · skills/adjudicating-taint-paths/SKILL.md · Decide whether a whitebox lead is a real bug by tracing taint from an untrusted source to a dangerous sink and…
git:20260816.faa9c07 · audit A · 26 stars
auditing-account-abstraction-and-paymaster-trust SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-account-abstraction-and-paymaster-trust/SKILL.md · Audit an ERC-4337 account-abstraction deployment for trust misplaced in the user-operation lifecycle: a smart account…
git:20260901.f8395d1 · audit A · 26 stars
auditing-account-recovery-and-reset-trust SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-account-recovery-and-reset-trust/SKILL.md · Audit password reset and account recovery flows for the trust that lets an attacker take over an account: a reset token…
git:20260829.b60f9b2 · audit A · 26 stars
auditing-admission-control-policy-gaps SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-admission-control-policy-gaps/SKILL.md · Audit cluster admission control for gaps that let a non-compliant or hostile workload through: a validating webhook…
git:20260828.ce55026 · audit A · 26 stars
auditing-ai-agent-permissions SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-ai-agent-permissions/SKILL.md · Audit what an AI agent is actually allowed to do versus what its task needs. Covers excessive agency (tools, scopes…
git:20260816.a6124f2 · audit A · 26 stars
auditing-android-component-exposure SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-android-component-exposure/SKILL.md · Audit an Android app for components another app on the device can reach and drive, after the manifest export flags and…
git:20260823.e797ed0 · audit A · 26 stars
auditing-android-intent-redirection-and-pendingintent SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-android-intent-redirection-and-pendingintent/SKILL.md · Audit Android privilege and access leaks through intent redirection and mutable pending intents, where a privileged…
git:20260908.e3ae18a · audit A · 26 stars
auditing-ansible-become-and-vault-trust SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-ansible-become-and-vault-trust/SKILL.md · Audit configuration-management privilege escalation and secret handling for trust that runs as root on every managed…
git:20260827.b5f41a4 · audit A · 26 stars
auditing-api-key-and-token-lifecycle SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-api-key-and-token-lifecycle/SKILL.md · Audit the lifecycle of API keys and access tokens for weaknesses that let one keep working past its intended bounds: a…
git:20260903.19a7174 · audit A · 26 stars
auditing-ble-and-gatt-authorization SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-ble-and-gatt-authorization/SKILL.md · Audit a Bluetooth Low Energy device for missing authorization on its GATT attributes: a characteristic performing a…
git:20260901.f8395d1 · audit A · 26 stars
auditing-break-glass-account-trust SKILL.md skill
unboundcompute/security-agent-skills · skills/auditing-break-glass-account-trust/SKILL.md · Audit emergency break-glass and privileged-access accounts for the ways their standing power outlives the emergency…
git:20260903.19a7174 · audit A · 26 stars

Every file in unboundcompute/security-agent-skills

Browse by kind, by grade A, or by owner.