RIPDPI: Skill for Claude Code

.agents/skills/desync-engine/SKILL.md

desync-engine is a skill for Claude Code, Codex from po4yka/RIPDPI. It costs 61 tokens per session (7,775 once invoked), scanned A, original, BSD-3-Clause.

A technical reference for a network traffic pipeline that changes how connection data appears to inspection systems while keeping it valid for the destination server. DPI, or deep packet inspection, examines network packets to identify protocols and destinations.

In plain words
What is it for?
Use it when changing TCP or UDP processing, packet offsets, fake packets, time-to-live settings, TLS setup data, or desynchronisation techniques in the RIPDPI project.
Why use it?
It gives developers the background needed to understand or change packet splitting, ordering, fake packets, and related connection behavior without losing track of the pipeline's layers.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is po4yka/RIPDPI's own configuration. It tells Claude Code and Codex how to work on RIPDPI itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything RIPDPI configures →

Reuse

Borrowing it

Nothing to install: this file belongs to po4yka/RIPDPI. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/po4yka/RIPDPI/main/.agents/skills/desync-engine/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/po4yka/RIPDPI

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for desync-engine

README.md
[![agentmods](https://agentmods.dev/badge/skills/po4yka/ripdpi/desync-engine/github.svg)](https://agentmods.dev/skills/po4yka/ripdpi/desync-engine)
Your own site
<a href="https://agentmods.dev/skills/po4yka/ripdpi/desync-engine"><img src="https://agentmods.dev/badge/skills/po4yka/ripdpi/desync-engine/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for desync-engine

Your own site · 80×15
<a href="https://agentmods.dev/skills/po4yka/ripdpi/desync-engine"><img src="https://agentmods.dev/badge/skills/po4yka/ripdpi/desync-engine.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 61 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 7,775 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00061 $0.07775
Opus 5.5 $0.00024 $0.03110
Sonnet 5.5 $0.00012 $0.01555
Haiku 4.5 $0.00006 $0.00777

Measured yesterday against content hash 6ecdd24cfbf3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

desync-engine scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.agents/skills/desync-engine/SKILL.md · 458 lines

How it starts

The opening of the file, as written. The whole thing — 458 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Desync Engine -- RIPDPI

Purpose

Document the three-layer DPI desync pipeline that RIPDPI uses to evade middlebox inspection: configuration model (ripdpi-config), planning (ripdpi-desync), and execution (ripdpi-desync-runtime through ripdpi-proxy-runtime). Apply this skill when modifying any layer of the pipeline, adding new desync techniques, or reasoning about why a strategy works (or doesn't) on a specific Android deployment mode.

Conceptual model

DPI (Deep Packet Inspection) systems like Russia's middlebox inspect the first packets of a TCP/UDP connection to fingerprint the protocol and destination. They look for TLS ClientHello SNI fields, HTTP Host headers, and QUIC Initial packet metadata. If the destination matches a blocklist, the DPI injects a RST or drops the connection.

Desync evasion works by making the first packets unreadable to the DPI while remaining valid to the destination server. Techniques include:

  • Splitting the payload across multiple TCP segments so no single packet contains the full SNI/Host value.
  • Disordering segments so the DPI reassembly engine sees garbage first.
  • Injecting fake packets with low TTL that reach the DPI but expire before the destination, poisoning the DPI's reassembly buffer.
  • TLS record fragmentation to split the ClientHello across TLS records within a single TCP segment.
  • UDP/QUIC fake bursts and SNI splitting for QUIC Initial packets.

Architecture overview

The desync engine is a three-layer pipeline:

Layer 1: Configuration model (ripdpi-config)

Files: native/rust/crates/ripdpi-config/src/model/ (start at mod.rs; group and runtime settings are split by domain)

Defines the data model: DesyncGroup, DesyncGroupActionSettings, TcpChainStep, TcpChainStepKind, UdpChainStep, UdpChainStepKind, OffsetBase, OffsetExpr, ActivationFilter, DesyncMode, EntropyMode, QuicFakeProfile, SeqOverlapFakeMode. These are pure data types with no I/O -- they represent what the user configured.

Read the full file on GitHub · 458 lines

Files

What ships with it

2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 458 lines · 61 tokens per session scan A 6ecdd24cfbf3

Subscribe to this mod's changes

desync-engine is a skill published in the GitHub repository po4yka/RIPDPI (85 stars, last pushed yesterday), licensed BSD-3-Clause. It adds 61 tokens to every session and 7,775 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-07.

Related

desync-engine compared with skills from other repositories

desync-engine next to 6 skills from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
desync-enginepo4yka/RIPDPISkill85yesterdayBSD-3-ClauseA61Claude Code, Codex
cindy-plugin-creatormakecindy/cindySkill9.5ktodayApache-2.0A48Codex
argent-react-native-app-workflowsoftware-mansion/argentSkill3.2ktodayApache-2.0A46Claude Code, Codex
engineer-bot-opsmakecindy/cindySkill9.5ktodayApache-2.0A57Claude Code, Codex
apple-search-ads-review-evidencemakecindy/cindySkill9.5ktodayApache-2.0A59Claude Code, Codex
seo-aeo-desk-evidencemakecindy/cindySkill9.5ktodayApache-2.0A43Claude Code, Codex
argent-tv-interactsoftware-mansion/argentSkill3.2ktodayApache-2.0A107Claude Code, Codex