forked from Keelrleek/Harbor
Every agent reaches GhidrAssistMCP through Harbor's own MCP server, so no per-host configuration is needed. Configured by ghidra_url in [reverse]; long decompiles are polled; also 'harbor re ghidra <tool> <json>'.
120 lines
5.9 KiB
Python
120 lines
5.9 KiB
Python
"""The shared reverse-engineering lookups, against a small synthetic image and annotation set."""
|
|
import json
|
|
import shutil
|
|
import struct
|
|
import unittest
|
|
|
|
from helpers import temp_dir
|
|
from harbor.reverse import Reverse
|
|
from harbor.util import HarborError
|
|
|
|
|
|
class ReverseTools(unittest.TestCase):
|
|
def setUp(self):
|
|
self.d = temp_dir()
|
|
base = 0x401000
|
|
image = bytearray(0x100)
|
|
# 0x401000: mov ecx,[0x401080]; call 0x401040; fmul dword [0x401090]; ret
|
|
code = bytes.fromhex("8b0d80104000") + b"\xe8" + struct.pack("<i", 0x401040 - 0x40100B) \
|
|
+ bytes.fromhex("d80d90104000") + b"\xc3"
|
|
image[0:len(code)] = code
|
|
image[0x40] = 0xC3 # 0x401040: ret
|
|
image[0x90:0x94] = struct.pack("<f", 100.0) # 0x401090: 100.0f
|
|
image[0xA0:0xA5] = b".cue\0"
|
|
(self.d / "image.bin").write_bytes(bytes(image))
|
|
facts = [{"address": "0x401000", "size": len(code), "name": "FUN_00401000", "calls": ["0x401040"],
|
|
"data": ["0x401080"], "imports": [], "strings": []},
|
|
{"address": "0x401040", "size": 1, "name": "FUN_00401040", "calls": [], "data": [],
|
|
"imports": [], "strings": []}]
|
|
(self.d / "facts.jsonl").write_text("\n".join(json.dumps(f) for f in facts), encoding="utf-8")
|
|
ann = self.d / "annotations"
|
|
ann.mkdir()
|
|
(ann / "names.csv").write_text("address,name,category,note\n0x00401000,GetLevel,game,\"Reads it.\"\n"
|
|
"0x00401040,Helper,game,x\n", encoding="utf-8")
|
|
(ann / "globals.csv").write_text("address,name,type,note\n0x00401080,g_root,int,x\n", encoding="utf-8")
|
|
(ann / "functions.csv").write_text("address,signature,note\n", encoding="utf-8")
|
|
(ann / "prototypes.h").write_text("int GetLevel(void); /* plain ret */\n", encoding="utf-8")
|
|
self.r = Reverse({"image": str(self.d / "image.bin"), "base": "0x401000", "facts": str(self.d / "facts.jsonl"),
|
|
"annotations": str(ann)})
|
|
|
|
def tearDown(self):
|
|
shutil.rmtree(self.d, ignore_errors=True)
|
|
|
|
def test_disassembly_names_calls_globals_and_constants(self):
|
|
text = self.r.disassemble("GetLevel")
|
|
self.assertIn("0x401080 g_root", text)
|
|
self.assertIn("0x401040 Helper", text)
|
|
self.assertIn("= 100.0f", text)
|
|
|
|
def test_xrefs_bytes_and_lookup(self):
|
|
x = self.r.xrefs("0x401040")
|
|
self.assertEqual([c["name"] for c in x["callers"]], ["GetLevel"])
|
|
self.assertEqual(self.r.bytes_at("0x401090", 4, "float")["float"], ["100.0"])
|
|
self.assertEqual(self.r.bytes_at("0x4010A0", as_type="string")["string"], ".cue")
|
|
info = self.r.lookup("GetLevel")
|
|
self.assertIn("plain ret", info["prototype"])
|
|
with self.assertRaises(HarborError):
|
|
self.r.bytes_at("0x500000")
|
|
with self.assertRaises(HarborError):
|
|
Reverse({})
|
|
|
|
|
|
class GhidraForwarder(unittest.TestCase):
|
|
"""The ghidra() tool against a fake streamable-HTTP MCP server: a session, a polled async task, a refusal."""
|
|
|
|
def setUp(self):
|
|
import http.server
|
|
import threading
|
|
calls = self.calls = []
|
|
|
|
class Fake(http.server.BaseHTTPRequestHandler):
|
|
def log_message(self, *a):
|
|
pass
|
|
|
|
def do_POST(self):
|
|
body = json.loads(self.rfile.read(int(self.headers["Content-Length"])))
|
|
calls.append((body["method"], body.get("params", {}).get("name"), self.headers.get("Mcp-Session-Id")))
|
|
if "id" not in body:
|
|
self.send_response(202)
|
|
self.end_headers()
|
|
return
|
|
name = body.get("params", {}).get("name")
|
|
if body["method"] == "initialize":
|
|
result = {"protocolVersion": "2025-06-18"}
|
|
elif body["method"] == "tools/list":
|
|
result = {"tools": [{"name": "get_code", "description": "Code",
|
|
"inputSchema": {"properties": {"function": {}}, "required": ["function"]}}]}
|
|
elif name == "get_code":
|
|
result = {"content": [{"type": "text", "text": "Task submitted for async execution.\n\n"
|
|
"Task ID: 11111111-2222-3333-4444-555555555555\nStatus: RUNNING"}]}
|
|
elif name == "get_task_status":
|
|
result = {"content": [{"type": "text", "text": "[Context] Operating on: image.bin\n\nint F(void)"}]}
|
|
else:
|
|
result = {"isError": True, "content": [{"type": "text", "text": "Refused: read-only"}]}
|
|
data = ("data: " + json.dumps({"jsonrpc": "2.0", "id": body["id"], "result": result}) + "\n\n").encode()
|
|
self.send_response(200)
|
|
self.send_header("Content-Type", "text/event-stream")
|
|
self.send_header("Mcp-Session-Id", "s1")
|
|
self.send_header("Content-Length", str(len(data)))
|
|
self.end_headers()
|
|
self.wfile.write(data)
|
|
|
|
self.server = http.server.ThreadingHTTPServer(("127.0.0.1", 0), Fake)
|
|
threading.Thread(target=self.server.serve_forever, daemon=True).start()
|
|
from harbor.reverse import Ghidra
|
|
self.g = Ghidra({"ghidra_url": f"http://127.0.0.1:{self.server.server_port}/mcp"})
|
|
|
|
def tearDown(self):
|
|
self.server.shutdown()
|
|
self.server.server_close()
|
|
|
|
def test_tools_call_and_refusal(self):
|
|
self.assertEqual(self.g.tools()[0]["required"], ["function"])
|
|
self.assertEqual(self.g.call("get_code", {"function": "F"}), "int F(void)")
|
|
self.assertIn(("tools/call", "get_task_status", "s1"), self.calls)
|
|
with self.assertRaises(HarborError):
|
|
self.g.call("rename_symbol", {})
|
|
with self.assertRaises(HarborError):
|
|
from harbor.reverse import Ghidra
|
|
Ghidra({"ghidra_url": "http://127.0.0.1:9/mcp"}).tools()
|