Kkrpc Interop

What it does

When your agent uses it

• Implementing kkrpc clients
• Servers in non-TypeScript languages
• Speaking the stable compact protocol
• Reference implementations in Go

Example prompts

• “/kkrpc-interop”

Requirements

• Python 3
• Compatibility (from SKILL.md): Works with any language that can parse JSON and implement stdio/WebSocket transports

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

1. 1Convert a dotted method name such as math.add to p: ["math", "add"].
2. 2Encode arguments, replacing callables with callback marker objects.
3. 3Send a request record with t: "q", unique id, op, p, and optional a or v.
4. 4Store a pending completion keyed by request ID.
5. 5In the read loop, resolve pending calls for t: "r" records.
6. 6In the read loop, dispatch t: "cb" records to stored callbacks.

What it can do on your machine

Read from SKILL.md and the folder at commit e558873. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  No scripts in the folder and no shell commands in SKILL.md (its code samples are json and python).

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

• Compatibility

  Works with any language that can parse JSON and implement stdio/WebSocket transports

  From compatibility in the SKILL.md frontmatter.

Context cost

Kkrpc Interop loads about 1.8k tokens when it runs. Until then it costs about 50 tokens; SKILL.md has 644 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~50

When it runs · the whole SKILL.md, loaded when a task matches

~1.8k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from kunkunsh/kkrpc at commit e558873, republished under its MIT licence (© kunkunsh). 644 words, ~1,765 tokens.

name

kkrpc-interop

description

Use when implementing kkrpc clients or servers in non-TypeScript languages, speaking the stable compact protocol, transports, and reference implementations in Go, Python, Rust, or Swift.

compatibility

Works with any language that can parse JSON and implement stdio/WebSocket transports

version

1.0.0

license

MIT

metadata.author

kkrpc

metadata.domain

cross-language-rpc

metadata.tags

rpc, interop, protocol, go, python, rust, swift

{
	"t": "q",
	"id": "a1b2-c3d4",
	"op": "call",
	"p": ["math", "add"],
	"a": [1, 2]
}

{
	"t": "r",
	"id": "a1b2-c3d4",
	"v": 3
}

{
	"t": "r",
	"id": "a1b2-c3d4",
	"e": {
		"n": "Error",
		"m": "Division by zero",
		"s": "optional stack"
	}
}

{
	"t": "cb",
	"id": "callback-id",
	"a": ["progress", 50]
}

{
	"__kkrpc_next_arg__": "callback",
	"id": "callback-id"
}

{
	"__kkrpc_next_arg__": "value",
	"v": "actual value"
}

import uuid

def generate_uuid() -> str:
    return str(uuid.uuid4())

read() -> string?   # Read one message, null/None if closed
write(message)      # Write one JSON string
close()             # Close connection

def call(method, *args):
    request_id = generate_uuid()
    payload = {
        "t": "q",
        "id": request_id,
        "op": "call",
        "p": method.split("."),
        "a": [encode_arg(arg) for arg in args],
    }
    transport.write(json.dumps(payload))
    response = wait_for_response(request_id)
    if "e" in response:
        raise RpcError(response["e"].get("m", "RPC error"))
    return response.get("v")

def handle(message):
    if message.get("t") != "q":
        return
    try:
        method = ".".join(message.get("p", []))
        args = [decode_arg(arg) for arg in message.get("a", [])]
        result = api[method](*args)
        transport.write(json.dumps({"t": "r", "id": message["id"], "v": result}))
    except Exception as exc:
        transport.write(json.dumps({
            "t": "r",
            "id": message["id"],
            "e": {"n": exc.__class__.__name__, "m": str(exc)},
        }))

© kunkunsh, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/interop of kunkunsh/kkrpc.

Open the folder on GitHubat commit e558873