agentsclimarketplace

Reverse engineering custom c2 protocols

Skill meltedinhex/analyst-ai-pack/skills/reverse-engineering-custom-c2-protocols

Reverses proprietary command-and-control protocols: locating send/recv routines, recovering the message framing and encryption/encoding, and reconstructing the command set to build a decoder or emulator. Activates for requests to reverse a custom C2 protocol, decode beacon traffic, or document a malware command structure.From its SKILL.md

Install
npx -y skills add meltedinhex/analyst-ai-pack --skill reverse-engineering-custom-c2-protocols

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 21 stars21 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its file declares

Copied from the file, not written here

The file declares its own license as Apache-2.0. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.

SKILL.md

3.3 KB, 569 tokens by cl100k_base, as published. Nobody here has run it

Reverse Engineering Custom C2 Protocols

When to Use

  • A sample uses a non-standard protocol (or a custom layer over HTTP/TCP) and you must decode its traffic.
  • You need to recover message framing, the encryption/encoding scheme, and the command set.
  • You want to build a standalone decoder or emulator to parse captured beacon traffic.

Do not use this for documented protocols handled by existing dissectors — use the protocol analyzer directly instead of reversing from scratch.

Prerequisites

  • A disassembler/debugger and a network capture (PCAP) of the sample's traffic where possible.
  • The crypto/obfuscation skills for recovering keys and encodings.

Workflow

Step 1: Locate the network routines

Find send/recv/WSASend/WinHTTP calls (or the framework wrappers) and the buffers they operate on. These bracket the serialization and crypto.

Step 2: Recover framing

Trace how outbound buffers are built: magic bytes, length prefixes, sequence/ID fields, and type/opcode fields. Document the header layout.

[ magic(4) ][ len(4 LE) ][ opcode(1) ][ flags(1) ][ payload(len) ]

Step 3: Recover crypto/encoding

Identify the transform applied before send / after recv (XOR with embedded key, RC4, AES, base64). Recover keys from the binary or from the key-setup routine.

Step 4: Reconstruct the command set

Map opcodes to handlers (shell, download, upload, sleep/jitter, exit) by following the dispatch switch. Build a table of command → behavior.

Step 5: Build a decoder and validate

Implement a parser/decryptor and run it against captured traffic; confirm it yields coherent commands and recovers the beacon's config.

python scripts/analyst.py decode capture.bin --key 0x5a --magic 4d5a4331

Validation

  • The decoder parses every message in a real capture without desync.
  • Decrypted payloads are coherent (printable commands / structured config).
  • The opcode table matches the dispatch logic observed in the binary.

Pitfalls

  • Assuming fixed keys when the protocol negotiates a session key in the first exchange.
  • Missing a length/sequence field and desynchronizing after the first message.
  • Confusing transport (TLS) with the inner custom layer; decode the inner layer.

References

What ships with it: 3 files

3.5 KB alongside SKILL.md, 1 of them executable

references/

scripts/

Keep looking

Skills are one crate of 326,144. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.