Write Vfunc As YAML

Write Virtual Function IDA Analysis Output as YAML

Persist virtual function analysis results to a YAML file beside the binary using IDA Pro MCP.

Prerequisites

Before using this skill, you should have:

1. Identified and renamed the target virtual function
2. Generated a unique signature using /generate-signature-for-function
3. Obtained vtable information using /get-vtable-index

Required Parameters

Optional Parameters

When func_addr is None, the following fields will be omitted from output: func_va, func_rva, func_size. When func_sig is None, the func_sig field will be omitted from output. When vfunc_sig is None, the vfunc_sig field will be omitted from output.

Method

mcp__ida-pro-mcp__py_eval code="""
import idaapi
import os
import yaml

# === REQUIRED: Replace these values ===
func_name = "<func_name>"           # e.g., "CCSPlayerController_ChangeTeam"
# ======================================

# === OPTIONAL: Set to None to omit from output ===
func_addr = <func_addr>             # e.g., 0x180999830 or None
func_sig = <func_sig>               # e.g., "48 89 5C 24 08" or None
vfunc_sig = <vfunc_sig>               # e.g., "FF 90 80 04 00 00 4C 8B AC 24 ?? ?? ?? ??" or None
# =================================================

# === VTABLE INFO: Replace these values ===
vtable_name = "<vtable_name>"       # e.g., "CCSPlayerController"
vfunc_offset = <vfunc_offset>       # e.g., 0x330
vfunc_index = <vfunc_index>         # e.g., 102
# =========================================

# Get binary path and determine platform
input_file = idaapi.get_input_file_path()
dir_path = os.path.dirname(input_file)

if input_file.endswith('.dll'):
    platform = 'windows'
    image_base = idaapi.get_imagebase()
else:
    platform = 'linux'
    image_base = 0x0

# Build data dictionary conditionally
data = {}

data['func_name'] = func_name

if func_addr is not None:
    func = idaapi.get_func(func_addr)
    func_size = func.size() if func else 0
    func_rva = func_addr - image_base
    data['func_va'] = hex(func_addr)
    data['func_rva'] = hex(func_rva)
    data['func_size'] = hex(func_size)

if func_sig is not None:
    data['func_sig'] = func_sig

if vfunc_sig is not None:
    data['vfunc_sig'] = vfunc_sig

data['vtable_name'] = vtable_name
data['vfunc_offset'] = hex(vfunc_offset)
data['vfunc_index'] = vfunc_index

yaml_path = os.path.join(dir_path, f"{func_name}.{platform}.yaml")
with open(yaml_path, 'w', encoding='utf-8') as f:
    yaml.dump(data, f, default_flow_style=False, sort_keys=False, allow_unicode=True)
print(f"Written to: {yaml_path}")
"""

Output File Naming Convention

The output YAML filename follows this pattern:

• <func_name>.<platform>.yaml

Examples:

• server.dll → CCSPlayerController_ChangeTeam.windows.yaml
• server.so / libserver.so → CCSPlayerController_ChangeTeam.linux.yaml

Output YAML Format

Full output (with func_addr and func_sig provided):

func_name: CCSPlayerController_ChangeTeam
func_va: 0x180999830      # Virtual address - changes with game updates (optional)
func_rva: 0x999830        # Relative virtual address (VA - image base) - changes with game updates (optional)
func_size: 0x301          # Function size in bytes - changes with game updates (optional)
func_sig: 48 89 5C 24 08  # Unique byte signature (optional)
vtable_name: CCSPlayerController
vfunc_offset: 0x330       # Offset from vtable start - changes with game updates
vfunc_index: 102          # vtable[102] - changes with game updates

Minimal output (with func_addr=None, func_sig=None, vfunc_sig=None):

func_name: CCSPlayerController_ChangeTeam
vtable_name: CCSPlayerController
vfunc_offset: 0x330       # Offset from vtable start - changes with game updates
vfunc_index: 102          # vtable[102] - changes with game updates

Platform Detection

The skill automatically detects the platform based on file extension:

• .dll → Windows (uses idaapi.get_imagebase() for image base)
• .so → Linux (uses 0x0 as image base)

Notes

• All values marked "changes with game updates" should be regenerated when analyzing new binary versions
• The YAML file is written to the same directory as the input binary
• When func_addr is provided, func_size is automatically calculated from IDA's function analysis
• When func_addr is provided, func_rva is automatically calculated as func_va - image_base
• When func_addr / func_sig / vfunc_sig is None, those fields are omitted from the output entirely
• This skill is specifically for virtual functions that have vtable information
• For regular functions without vtable, use /write-func-as-yaml instead